Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2417
Encontrada(s) 25 página(s)

DIGITAL.INTA Repositorio Digital del Instituto Nacional de Técnica Aeroespacial
Publication Contribución de congreso (ConferenceOutput)

Round-off noise estimation of fixed-point algorithms using Modified Affine Arithmetic and Legendre Polynomials

  • Esteban, L.
  • López Martín, E.
  • Regadío, A.
Conference Location: Segovia, Spain, The implementation of algorithms in fixed-point format causes the apparition of Round-Off Noise which propagates through the different functional units of the system. This issue causes the Signal-to-Noise Ratio of the outputs is degraded. Given an algorithm, it is essential to estimate the integer and fractional bit-widths of all the variables and operations to comply with the Signal-to-Noise Ratio requirements. In this context, Affine Arithmetic can obtain fast and accurate estimations of the bit-widths for linear systems. However, for non-linear systems, Affine Arithmetic loses the temporal correlation of the variables. Other existing frameworks are either time consuming or lead to inaccurate bound estimations. In this paper, a Modified Affine Arithmetic framework with Legendre polynomials is used to obtain fast and accurate bound estimations also for non-linear systems. Moreover, the approach proposed in this paper obtains speedups in the range of 7 to 100 compared to Monte-Carlo simulations, This work has been partially supported by the spanish MINECO under projects AYA2015-65973-C3-3-R and RTI2018-096228-B-C33
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Conditional dynamic partial order reduction and optimality results

  • Isabel Márquez, Miguel
Testing concurrent systems requires exploring all possible nondeterministic interleavings that the concurrent execution may have, as any of the interleavings may reveal an erroneous behaviour of the system. This introduces a combinatorial explosion on the number of states that must be considered, which leads often to a computationally intractable problem. In the present PhD thesis, this challenge will be addressed through the development of new Partial Order Reduction techniques (POR). The cornerstone of POR theory is the notion of independence, that is used to decided whether each pair of concurrent events p and t are in a race and thus both executions p · t and t · p must be explored. A fundamental goal of this thesis is to introduce notions of conditional independence –which ensure the commutativity of the considered events p and t under certain conditions that can be evaluated in the explored state– with a DPOR algorithm in order to alleviate the combinatorial explosion problem. The new techniques that we propose in the thesis have been implemented within the SYCO tool. We have carried out accompanying experimental evaluations to prove the effectiveness and applicability of the proposed techniques. Finally, we have successfully verified a range of properties for several case studies of Software-Defined Networks to illustrate the potential of the approach, scaling to larger networks than related techniques.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Enriched Semantic Graphs for Extractive Text Summarization

  • García Sevilla, Antonio Fernando
  • Fernández-Isabel, Alberto
  • Díaz Esteban, Alberto
Automatic extraction of semantic information from unstructured text has always been an important goal of natural language processing. While the best structure for semantic information is still undecided, graphbased representations enjoy a healthy following. Some of these representations are extracted directly from the text and external knowledge, while others are built from linguistic insight, created from the deep analysis of the surface text. In this document a combination of both approaches is outlined, and its application for extractive text summarization is described. A pipeline for this task has been implemented, and its results evaluated against a collection of documents from the DUC2003 competition. Graph construction is fully automatic, and summary creation is based on the clustering of conceptual nodes. Different configurations for the semantic graphs are used and compared, and their fitness for the task discussed.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Constrained Dynamic Partial Order Reduction

  • Albert Albiol, Elvira María
  • Gómez-Zamalloa Gil, Miguel
  • Isabel Márquez, Miguel
  • Rubio, Albert
The cornerstone of dynamic partial order reduction (DPOR) is the notion of independence that is used to decide whether each pair of concurrent events p and t are in a race and thus both p · t and t · p must be explored. We present constrained dynamic partial order reduction (CDPOR), an extension of the DPOR framework which is able to avoid redundant explorations based on the notion of conditional independence— the execution of p and t commutes only when certain independence constraints (ICs) are satisfied. ICs can be declared by the programmer, but importantly, we present a novel SMT-based approach to automatically synthesize ICs in a static pre-analysis. A unique feature of our approach is that we have succeeded to exploit ICs within the state-of-the-art DPOR algorithm, achieving exponential reductions over existing implementations.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

LFOC: A Lightweight Fairness-Oriented Cache Clustering Policy for Commodity Multicores

  • García García, Adrián
  • Sáez Alcaide, Juan Carlos
  • Castro Rodríguez, Fernando
  • Prieto Matías, Manuel
Multicore processors constitute the main architecture choice for modern computing systems in different market segments. Despite their benefits, the contention that naturally appears when multiple applications compete for the use of shared resources among cores, such as the last-level cache (LLC), may lead to substantial performance degradation. This may have a negative impact on key system aspects such as throughput and fairness. Assigning the various applications in the workload to separate LLC partitions with possibly different sizes, has been proven effective to mitigate shared-resource contention effects. In this article we propose LFOC, a clustering-based cache partitioning scheme that strives to deliver fairness while providing acceptable system throughput. LFOC leverages the Intel Cache Allocation Technology (CAT), which enables the system software to divide the LLC into different partitions. To accomplish its goals, LFOC tries to mimic the behavior of the optimal cache-clustering solution, which we could approximate by means of a simulator in different scenarios. To this end, LFOC effectively identifies streaming aggressor programs and cache sensitive applications, which are then assigned to separate cache partitions. We implemented LFOC in the Linux kernel and evaluated it on a real system featuring an Intel Skylake processor, where we compare its effectiveness to that of two state-of-the-art policies that optimize fairness and throughput, respectively. Our experimental analysis reveals that LFOC is able to bring a higher reduction in unfairness by leveraging a lightweight algorithm suitable for adoption in a real OS.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

EMINENT: Embarrassingly parallel mutation testing

  • Cerro Cañizares, Pablo
  • Núñez Covarrubias, Alberto
  • García Merayo, María De Las Mercedes
During the last decade, the fast evolution in communication networks has facilitated the development of complex applications that manage vast amounts of data, like Big Data applications. Unfortunately, the high complexity of these applications hampers the testing process. Moreover, generating adequate test suites to properly check these applications is a challenging task due to the elevated number of potential test cases. Mutation testing is a valuable technique to measure the quality of the selected test suite that can be used to overcome this difficulty. However, one of the main drawbacks of mutation testing lies on the high computational cost associated to this process. In this paper we propose a dynamic distributed algorithm focused on HPC systems, called EMINENT, which has been designed to face the performance problems in mutation testing techniques. EMINENT alleviates the computational cost associated with this technique since it exploits parallelism in cluster systems to reduce the final execution time. In addition, several experiments have been carried out on three applications in order to analyse the scalability and performance of EMINENT. The results show that EMINENT provides an increase in the speed-up in most scenarios.
Docta Complutense
Publication Contribución de congreso (ConferenceOutput)

Metamorphic Testing of Oracle Quantum Programs

  • Abreu, Rui
  • Llana Díaz, Luis Fernando
  • Fernandes, Joao Paulo
  • Tavares, Guilherme
Quantum Computing is regarded with a growing sense of excitement and expectation. While the preliminary evidences of its practical interest accumulate, there are numerous challenges that must be addressed before quantum computing can reach generalized adoption and realize its full potential. One such challenge concerns assessing whether a quantum program is correct, namely in the sense that it implements the desired functionality. In the classical realm, program testing is a well-established approach to find defects in programs. While this approach should inspire quantum software development, porting classical testing techniques to the quantum world is by no means trivial. We propose a novel approach to test quantum programs. Our approach is based on the use of metamorphic relations, which allows us to overcome the well-known quantum measurement problem. We describe a series of exploratory experiments whose results provide evidence of theusefulness of our approach.
RIUMA. Repositorio Institucional de la Universidad de Málaga
Publication Contribución de congreso (ConferenceOutput)

ClinicAppChain: A Low-Cost Blockchain Hyperledger Solution for Healthcare.

  • Muñoz-Guerra, Daniel Jesús
  • Constantinescu, Denisa-Andreea
  • Asenjo-Plaza, Rafael
  • Fuentes-Fernández, Lidia
Work by Munoz and Fuentes is supported by the projects MAGIC P12-TIC1814 and HADAS TIN2015-64841-R (co-financed by FEDER funds). Work by Constantinescu and Asenjo is supported by the project TIN2016-80920-R, funded by the Spanish Government. ClinicAppChain has been supported by IMFAHE Foundation-Nodal Award of 2018. We are particularly grateful to other members of the awarded team that have guided us in the sanitary side and legal aspects: Esther Herrera, Laura Timanfaya, Patricia Rodríguez and Myriam Martínez., https://www.springernature.com/gp/open-science/policies/book-policies, Unified access to anonymous records through a trustworthy system is needed in our increasingly globalised world that still suffers from profoundly disconnected health care services. Emergencies are followed by redundant, costly and slow medical examinations without a global health data sharing mechanism. We establish the foundations of a decentralised healthcare ledger where patients decide what to share, with who, and with minimal costs. We review the state-of-the-art of transparent, auditable and interactive Blockchain-based healthcare studies, developing ClinicAppChain. Our solution features authentication, confidentiality and permissioned data sharing, considering the EU data protection regulation of 2018. ClinicAppChain is a cross-platform low-cost Blockchain prototype that empowers patients, hospitals, researchers, pharmaceuticals and insurance industries without crypto-currencies involved, and with a negligible energy foot-print (7.7 watts per node).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A novel approach for spectral efficiency analysis in MIMO cellular networks

  • George, Geordie
  • Mungara, Ratheesh K.
  • Lozano Solsona, Angel
  • Haenggi, Martin
Comunicació presentada a la 2017 IEEE International Conference on Communications (ICC), celebrada els dies 21 a 25 de maig de 2017 a París, França., The application of stochastic geometry to the analysis of wireless networks is shown to be propelled by (i) a clean separation of time scales, (ii) abstraction of small-scale effects via ergodicity, and (iii) an interference model reflecting the receiver's lack of knowledge of how individual interference terms are faded. These steps render the analysis simpler and more precise, and more amenable to incorporating subsequent features. Specifically, the paper presents easy-to-evaluate expressions for the ergodic spectral efficiency of cellular networks with single-user multiple-input multiple-output (MIMO)., This work was supported by Project TEC2015-66228-P (MINECO/FEDER, UE) and by the U.S. NSF through award CCF 1525904.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

MediaEval 2018 AcousticBrainz genre task: A baseline combining deep feature embeddings across datasets

  • Oramas, Sergio
  • Bogdanov, Dmitry
  • Porter, Alastair
Comunicació presentada al MediaEval 2018 Workshop celebrat a Sophia Antipolis (França) del 29 al 31 d'octubre de 2018., In this paper we present a baseline approach for the MediaEval 2018 AcousticBrainz Genre Task that takes advantage of stacking multiple feature embeddings learned on individual genre datasets by simple deep learning architectures. Although we employ basic neural networks, the combination of their deep feature embeddings provides a significant gain in performance compared to each individual network., This research has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements No 688382 (AudioCommons) and 770376-2 (TROMPA), as well as the Ministry of Economy and Competitiveness of the Spanish Government (Reference: TIN2015-69935-P).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Maker education in primary education: changes in students’ maker-mindset and gender differences

  • Martínez Moreno, Judit
  • Santos, Patricia
  • Hernández Leo, Davinia
Comunicació presentada a: TEEM'21 celebrat del 27 a 29 d'octubre de 2021 a Barcelona, Espanya., Students must learn how to be prepared to enter the global digital network and acquire skills for the 21st century. Maker education allows for the development of such skills. This paper focuses on the implementation of Maker activities within a primary education context through a 10-hour programme across 10 schools. The study aims to analyse the possible impacts of maker education interventions on students’ Maker mindset (self-efficacy, interest and motivation), as well as assessing potential gender differences regarding their prior experience with Maker education and changes experienced on their Maker mindset. Questionnaires were administered before and after the programme's implementation on a sample of 239 students. Results showed that Maker interventions can have positive effects on students’ Maker mindset, increasing their self-efficacy, interest and motivation. In terms of gender differences, girls seemed to have less prior experience with Maker education than boys, yet they reported the same level of interest and motivation. This study contributes to demonstrating the positive impact that Maker education can have on primary education school settings in order to develop 21st century digital skills and bridging the gender gap in STEAM-related subjects., This work has been partially funded by the EU Regional Development Fund and the National Research Agency of the Spanish Ministry of Science, under project grants TIN2017-85179-C3-3-R, PID2020-112584RB-C33, MDM-2015-0502, the Ramon y Cajal programme (P. Santos) and by ICREA under the ICREA Academia programme (D. Hernández-Leo, Serra Hunter).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Standard quasi-conformal flattening of the right and left atria

  • Nuñez-Garcia, Marta
  • Bernardino Perez, Gabriel
  • Doste Beltrán, Rubén
  • Zhao, Jichao
  • Camara, Oscar
  • Butakoff, Constantine
Two-dimensional standard representations of 3D anatomical structures are a simple and intuitive way for analysing patient information across populations and image modalities. They also allow convenient visualizations that can be included in clinical reports for a fast overview of the whole structure. While cardiac ventricles, especially the left ventricle, have an established standard representation (e.g. bull’s eye plot), the 2D depiction of the left (LA) and right atrium (RA) remains challenging due to their sub-structural complexity. Quasi-conformal flattening techniques, successfully applied to cardiac ventricles, require additional constraints in the case of the atria to correctly place the adjacent structures, i.e. the pulmonary veins, the vena cava (VC) or the appendages. Some registration-based methods exist to flatten the LA but they can be time-consuming and prone to errors if the geometries are very different. We propose a novel atrial flattening methodology where a quasi-conformal 2D map of both (left and right) atria is obtained quickly and without errors related to registration. In our approach the RA is mapped to a standard 2D map where the holes corresponding to superior and inferior VC are fixed within a disk. Similarly, the LA is divided into 5 regions which are then mapped to their analogous two-dimensional regions. We illustrate the application of the method to visualize atrial wall thickness measurements, and late gadolinium enhanced magnetic resonance data., This study was partially funded by the Spanish Ministry of Economy and Competitiveness (DPI2015-71640-R), by the “Fundació La Marató de TV3” (no 20154031) and by European Union Horizon 2020 Programme for Research and Innovation, under grant agreement No. 642676 (CardioFunXion).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Generalized planning as heuristic search

  • Segovia-Aguas, Javier
  • Jiménez, Sergio
  • Jonsson, Anders, 1973-
Comunicació presentada a ICAPS2021, celebrat del 2 a 13 d'agost de 2021 de manera virtual., Although heuristic search is one of the most successful approaches to classical planning, this planning paradigm does not apply straightforwardly to Generalized Planning (GP). Planning as heuristic search traditionally addresses the computation of sequential plans by searching in a grounded state-space. On the other hand, GP aims at computing algorithm-like plans that can branch and loop, and that generalize to a (possibly infinite) set of planning instances. This paper adapts the {\em planning as heuristic search} paradigm to the particularities of GP and presents the first native heuristic search approach to GP. First, the paper defines a novel GP solution space that is independent of the number of planning instances in a GP problem, and the size of these instances. Second, the paper defines several evaluation and heuristic functions, that do not require grounding, for guiding a combinatorial search in the GP solution space, making it possible to handle state variables with large numerical domains (e.g. integers). Lastly, the paper defines an algorithm for GP called Best-First Generalized Planning (BFGP), that implements a best-first search in the solution space guided by our evaluation/heuristic functions., Javier Segovia-Aguas is supported by TAILOR, a project funded by EU H2020 research and innovation programme no. 952215, an ERC Advanced Grant no. 885107, and grant TIN-2015-67959-P from MINECO, Spain. Sergio Jiménez is supported by the Ramon y Cajal program, RYC-2015-18009, the Spanish MINECO project TIN2017-88476-C2-1-R. Anders Jonsson is partially supported by Spanish grants PID2019-108141GB-I00 and PCIN-2017-082.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

CONNECT: CONtextual NamE disCovery for blockchain-based services in the IoT

  • Daza, Vanesa
  • Di Pietro, Roberto
  • Klimek, Ivan
  • Signorini, Matteo
Comunicació presentada a la 2017 IEEE International Conference on Communications (ICC), celebrada els dies 21 a 25 de maig de 2017 a Paris, França., The Internet of Things is gaining momentum thanks to the provided vision of seamlessly interconnected devices. However, a unified way to discover and to interact with the surrounding smart environment is missing. As an outcome, we have been assisting to the development of heterogeneous ecosystems, where each service provider adopts its own protocol- thus preventing IoT devices from interacting when belonging to different providers. And, the same is happening again for the blockchain technology which provides a robust and trusted way to accomplish tasks -unfortunately not providing interoperability thus creating the same heterogeneous ecosystems above highlighted. In this context, the fundamental research question we address is how do we find things or services in the Internet of Things. In this paper, we propose the first IoT discovery approach which provides an answer to the above question by exploiting hierarchical and universal multi-layered blockchains. Our approach does neither define new standards nor force service providers to change their own protocol. On the contrary, it leverages the existing and publicly available information obtained from each single blockchain to have a better knowledge of the surrounding environment. The proposed approach is detailed and discussed with the support of relevant use cases., This work is partly supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and Project TEC2015-66228-P (MINECO/FEDER,UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Can domain adaptation be handled as analogies?

  • Bel Rafecas, Núria
  • Pocostales, Joel
Comunicació presentada a: 11th International Conference on Language Resources and Evaluation, celebrada a Miyazaki, Japó, del 7 al 12 de maig del 2018., Aspect identification in user generated texts by supervised text classification might suffer degradation in performance when changing to other domains than the one used for training. For referring to aspects such as quality, price or customer services the vocabulary might differ and affect performance. In this paper, we present an experiment to validate a method to handle domain shifts when there is no available labeled data to retrain. The system is based on the offset method as used for solving word analogy problems in vector semantic models such as word embedding. Despite of the fact that the offset method indeed found relevant analogues in the new domain for the classifier initial selected features, the classifiers did not deliver the expected results. The analysis showed that a number of words were found as analogues for many different initial features. This phenomenon was already described in the literature as 'default words' or 'hubs'. However, our data showed that it cannot be explained in terms of word frequency or distance to the question word, as suggested., This work was supported by the Spanish TUNER project TIN2015-65308-C5-5-R (MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

End-to-end sound source separation conditioned on instrument labels

  • Slizovskaia, Olga
  • Kim, Leo
  • Haro Ortega, Gloria
  • Gómez Gutiérrez, Emilia, 1975-
Comunicació presentada a: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing celebrat de 12 al 17 de maig de 2019 a Brighton, Regne Unit., Can we perform an end-to-end music source separation with a variable number of sources using a deep learning model? This paper presents an extension of the Wave-UNet [1] model which allows end-to-end monaural source separation with a non-fixed number of sources. Furthermore, we propose multiplicative conditioning with instrument labels at the bottleneck of the Wave-U-Net and show its effect on the separation results. This approach can be further extended to other types of conditioning such as audio-visual source separation and score-informed source separation., This work has received funding from the Maria de Maeztu Programme (MDM-2015-0502), ERC Innovation Programme (grant 770376, TROMPA), and MINECO/FEDER UE project (TIN2015-70410-C2-1-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

What sentence are you referring to and why? Identifying cited sentences in scientific literature

  • AbuRa'ed, Ahmed Ghassan Tawfiq
  • Chiruzzo, Luis
  • Saggion, Horacio
Comunicació presentada a la International Conference Recent Advances in Natural Language Processing (RANLP 2017), celebrada els dies 2 a 8 de setembre de 2017 a Varna, Bulgària., In the current context of scientific information overload, text mining tools are of paramount importance for researchers who have to read scientific papers and assess their value. Current citation networks, which link papers by citation relationships (reference and citing paper), are useful to quantitatively understand the value of a piece of scientific work, however they are limited in that they do not provide information about what specific part of the reference paper the citing paper is referring to. This qualitative information is very important, for example, in the context of current community-based scientific summarization activities. In this paper, and relying on an annotated dataset of co-citation sentences, we carry out a number of experiments aimed at, given a citation sentence, automatically identify a part of a reference paper being cited. Additionally our algorithm predicts the specific reason why such reference sentence has been cited out of five possible reasons., This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Empirically investigating virtual learning companions to enhance social media literacy

  • Theophilou, Emily
  • Schwarze, Veronica
  • Börsting, Johanna
  • Sánchez Reina, Jesús Roberto
  • Scifo, Lidia
  • Lomonaco, Francesco
  • Aprin, Farbod
  • Ognibene, Dimitri
  • Taibi, Davide
  • Hernández Leo, Davinia
  • Eimler, Sabrina
Comunicació presentada a 4th International Conference on Higher Education Learning Methodologies and Technologies Online, celebrat del 21 al 23 de setembre de 2022 a Palerm, Itàlia., Social media platforms provide opportunities for users across the world to connect and communicate between them and engage into acts of social support and entertainment. Yet it can also bring negative consequences as it has been associated with poor mental health and life dissatisfaction. This underlines the importance of delivering social media literacy (SML) interventions that raise awareness of the dangers and threats that are hidden within. To this date, SML initiatives have shown their benefits towards the acquisition of SML skills through the forms of school interventions and mini-games. However, studies on promoting SML through social media platforms need to be also encouraged and innovative approaches to provide interactive scenarios with hands-on experiences need to be formulated. Hence, the project COURAGE introduces a new approach towards SML by proposing the integration of educational opportunities within a controlled social media platform. To provide students the opportunity to learn whilst they naturally explore social media we propose the integration of virtual learning companions. In this paper we report seven empirical approaches towards SML skills acquisition powered by virtual learning companions. The paper concludes with a discussion towards the benefits and limitations of using this type of SML interventions., This work has been partially funded by the National Research Agency of the Spanish Ministry (PID2020-112584RB-C33/MICIN/AEI/10.13039/501100011033, MDM-2015-0502) and the Volkswagen Stiftung (Courage, ref. 95 566). D. Hernández-Leo (Serra Húnter) acknowledges the support by ICREA under the ICREA Academia programme.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

LaSTUS/TALN+INCO @ CL-SciSumm 2018 - Using regression and convolutions for cross-document semantic linking and summarization of scholarly literature

  • AbuRa'ed, Ahmed Ghassan Tawfiq
  • Bravo Serrano, Àlex, 1984-
  • Chiruzzo, Luis
  • Saggion, Horacio
Comunicació presentada al congrés BIRNDL 2018, 3rd Joint Workshop on Bibliometric-enhanced Information Retrieval and Natural Language Processing for Digital Libraries que va tenir lloc el 21 de juliol de 2018 a Ann Arbor, Estats Units., In this paper we present several systems developed to partic- ipate in the 3rd Computational Linguistics Scienti c Document Summa- rization Shared challenge which addresses the problem of summarizing a scienti c paper taking advantage of its citation network (i.e., the pa- pers that cite the given paper). Given a cluster of scienti c documents where one is a reference paper (RP) and the remaining documents are papers citing the reference, two tasks are proposed: (i) to identify which sentences in the reference paper are being cited and why they are cited, and (ii) to produce a citation-based summary of the reference paper using the information in the cluster. Our systems are based on both supervised (Convolutional Neural Networks) and unsupervised techiques taking ad- vantage of word embeddings representations and features computed from the linguistic and semantic analysis of the documents., This work is (partly) supported by the Spanish Ministry of Economy and Com- petitiveness under the Maria de Maeztu Units of Excellence Programme (MDM- 2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

The MediaEval 2018 AcousticBrainz genre task: content-based music genre recognition from multiple sources

  • Bogdanov, Dmitry
  • Porter, Alastair
  • Urbano, Julián
  • Schreiber, Hendrik
Comunicació presentada al MediaEval 2018 Workshop celebrat a Sophia Antipolis (França) del 29 al 31 d'octubre de 2018., This paper provides an overview of the AcousticBrainz Genre Task organized as part of the MediaEval 2018 Benchmarking Initiative for Multimedia Evaluation. The task is focused on content-based music genre recognition using genre annotations from multiple sources and large-scale music features data available in the AcousticBrainz database. The goal of our task is to explore how the same music pieces can be annotated differently by different communities following different genre taxonomies, and how this should be addressed by content-based genre recognition systems. We present the task challenges, the employed ground-truth information and datasets, and the evaluation methodology., We thank all contributors to AcousticBrainz. This research has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreements No 688382 (AudioCommons) and 770376-2 (TROMPA), as well as the Ministry of Economy and Competitiveness of the Spanish Government (Reference: TIN2015-69935-P). We also thank tagtraum industries for providing the Tagtraum genre annotations.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Multi-level mining and visualization of scientific text collections

  • Accuosto, Pablo
  • Ronzano, Francesco
  • Ferrés, Daniel
  • Saggion, Horacio
Comunicació presentada a: 6th International Workshop on Mining Scientific Publications (WOSP 2017), celebrat el 19 de juny a Toronto, Canada., We present a system to mine and visualize collections of scientific documents by semantically browsing information extracted from single publications or aggregated throughout corpora of articles. The text mining tool performs deep analysis of document collections allowing the extraction and interpretation of research paper’s contents. In addition to the extraction and enrichment of documents with metadata (titles, authors, affiliations, etc), the deep analysis performed comprises semantic interpretation, rhetorical analysis of sentences, triple-based information extraction, and text summarization. The visualization components allow geographicalbased exploration of collections, topic-evolution interpretation, and collaborative network analysis among others. The paper presents a case study of a bilingual collection in the field of Natural Language Processing (NLP)., This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and by the TUNER project (TIN2015- 65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

∃-STRIPS: existential quantification in planning and constraint satisfaction

  • Francès, Guillem
  • Geffner, Héctor
Comunicació presentada a la Twenty-Fifth International Joint Conference on Artificial Intelligence, celebrada els dies 9 a 15 de juliol de 2016 a Nova York., Existentially quantified variables in goals and action preconditions are part of the standard PDDL planning language, yet few planners support them, while those that do compile them away at an exponential cost. In this work, we argue that existential variables are an essential feature for representing and reasoning with constraints in planning, and that it is harmful to compile them away or avoid them altogether, since this hides part of the problem structure that can be exploited computationally. We show how to do this by formulating an extension of the standard delete-relaxation heuristics that handles existential variables. While this extension is simple, the consequences for both modeling and computation are important. Furthermore, by allowing existential variables in STRIPS and treating them properly, CSPs can be represented and solved in a direct manner as action-less,fluent-less STRIPS planning problems, something important for problems involving restrictions. In addition, functional fluents in Functional STRIPS can be compiled away with no effect on the structure and informativeness of the resulting heuristic. Experiments are reported comparing our native ∃-STRIPS planner with state-of-the-art STRIPS planners over compiled and propositional encodings, and with a Functional STRIPS planner., This work is partially supported by the M. de Maeztu Programme (MDM-2015-0502) and by grants TIN2015-67959 and CSD2010-00034, MEC, Spain.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Musical instrument recognition in user-generated videos using a multimodal convolutional neural network architecture

  • Slizovskaia, Olga
  • Gómez Gutiérrez, Emilia, 1975-
  • Haro Ortega, Gloria
Comunicació presentada a la International Conference on Multimedia Retrieval celebrada del 6 al 9 de juny de 2017 a Bucarest, Romania., This paper presents a method for recognizing musical instruments in user-generated videos. Musical instrument recognition from music signals is a well-known task in the music information retrieval (MIR) field, where current approaches rely on the analysis of the good-quality audio material. This work addresses a real-world scenario with several research challenges, i.e. the analysis of user-generated videos that are varied in terms of recording conditions and quality and may contain multiple instruments sounding simultaneously and background noise. Our approach does not only focus on the analysis of audio information, but we exploit the multimodal information embedded in the audio and visual domains. In order to do so, we develop a Convolutional Neural Network (CNN) architecture which combines learned representations from both modalities at a late fusion stage. Our approach is trained and evaluated on two large-scale video datasets: YouTube-8M and FCVID. The proposed architectures demonstrate state-of-the-art results in audio and video object recognition, provide additional robustness to missing modalities, and remains computationally cheap to train., is work is partly supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), the CASAS Spanish research project (TIN2015-70816-R), and project TIN2015-70410-C2- 1-R (MINECO/FEDER, UE). We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X GPU used for this research.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A plan for sustainable MIR evaluation

  • McFee, Brian
  • Humphrey, Eric J.
  • Urbano, Julián
The Music Information Retrieval Evaluation eXchange (MIREX) is a valuable community service, having established standard datasets, metrics, baselines, methodologies, and infrastructure for comparing MIR methods. While MIREX has managed to successfully maintain operations for over a decade, its long-term sustainability is at risk. The imposed constraint that input data cannot be made freely available to participants necessitates that all algorithms run on centralized computational resources, which are administered by a limited number of people. This incurs an approximately linear cost with the number of submissions, exacting significant tolls on both human and financial resources, such that the current paradigm becomes less tenable as participation increases. To alleviate the recurring costs of future evaluation campaigns, we propose a distributed, community-centric paradigm for system evaluation, built upon the principles of openness, transparency, reproducibility, and incremental evaluation. We argue that this proposal has the potential to reduce operating costs to sustainable levels. Moreover, the proposed paradigm would improve scalability, and eventually result in the release of large, open datasets for improving both MIR techniques and evaluation methods., B.M. is supported by the Moore-Sloan Data Science Environment at NYU. J.U. is supported by the Spanish Government: JdC postdoctoral fellowship, and projects TIN2015-70816-R and MDM-2015-0502.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Motion inpainting by an image-based geodesic AMLE method

  • Oliver, Maria
  • Raad, Lara
  • Ballester, Coloma
  • Haro Ortega, Gloria
Comunicació presentada al congrés 25th IEEE International Conference on Image Processing (ICIP) celebrat del 7 al 10 d'octubre de 2018 a Atenes, Grècia., This work presents an automatic method for optical flow inpainting. Given a video, each frame domain is endowed with a Riemannian metric based on the video pixel values. The missing optical flow is recovered by solving the Absolutely Minimizing Lipschitz Extension (AMLE) partial differential equation on the Riemannian manifold. An efficient numerical algorithm is proposed using eikonal operators for nonlinear elliptic partial differential equations on a finite graph. The choice of the metric is discussed and the method is applied to optical flow inpainting and sparse-to-dense optical flow estimation, achieving top-tier performance in terms of End-Point-Error (EPE)., This work is partially supported by MINECO/FEDER UE project, with reference TIN2015-70410-C2-1-R and by RISE project NoMADS.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Data augmentation for deep learning source separation of HipHop songs

  • Martel Baro, Héctor
  • Miron, Marius
Comunicació presentada a: 10th International Workshop on Machine Learning and Music, celebrat el 6 d'octbure de 2017 a Barcelona., Training deep learning source separation methods involves computationally intensive procedures relying on large multi-track datasets. In this paper we use data augmentation to improve hip hop source sepa- ration using small training datasets. We analyze different training strate- gies and data augmentation techniques with respect to their generaliza- tion capabilities. Moreover, we propose a hip hop multi-track dataset and we implemented a web demo to demonstrate our use scenario. The evaluation is done on a part of the dataset and hip-hop songs from an external dataset.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Distributed retrieval practice enhances primary school students’ retention of computational thinking concepts

  • Casanova de Vilalta, Lydia
  • Theophilou, Emily
  • Beardsley, Marc Yoshimi
  • Santos Rodríguez, Patrícia
  • Hernández Leo, Davinia
Comunicació presentada al 17th International Conference on Cognition and Exploratory Learning in Digital Age (CELDA 2020), celebrat del 18 al 20 de novembre de 2020 a Lisboa, Portugal., Efforts are being made to add Computational Thinking (CT) to the curriculum of primary schools in order to develop student abilities to use computational tools to solve problems and better prepare them for an increasingly digital society. Distributed Practice (DP) and Retrieval Practice (RP) are evidence-based practices that have shown to lead to efficient and durable learning. Studies investigating Distributed Retrieval Practice (DRP), a combination of the two evidence-based practices, have found positive effects of DRP on university student learning of anatomy and vocabulary items. As CT is considered to be a complex thinking skill, one that is acquired over time with practice, we investigated whether DRP had durable effects on primary student learning of CT concepts. We conducted a quasi-experimental study involving 20 primary school students between 11 and 12 years of age. The students participated in 6 weekly sessions on CT and completed two tests – one at the end of the fifth session (immediate) and the other a week later (delayed). At the end of each of the first 4 sessions, students performed DRP by completing a class-wide review quiz consisting of 5 multiple-choice questions. The results of the study show that students attained higher scores on the delayed test (72.6% ± 19.2) in comparison to the immediate test (67.9% ± 20.7). However, the difference in scores was not significant. The results suggest that DRP may play a role in helping students overcome the natural decay of memory, however, studies with a larger sample size are required. Further, this study demonstrates that DRP can be applied in an authentic classroom environment that involves primary school student learning of a complex subject such as CT., This work has been partially funded by the European Regional Development Fund, Erasmus+, and the National Research Agency of the Spanish Ministry of Science, Innovation and Universities under project grants TIN2017-85179-C3-3-R, 2017-1-ES01-KA201-038220 and MDM-2015-0502. Dr. Hernández-Leo acknowledges the support by ICREA under the ICREA Academia programme.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Designing fully secure protocols for secure two-party computation of constant-domain functions

  • Daza, Vanesa
  • Makriyannis, Nikolaos
Comunicació presentada a la TCC 2017 Theory of Cryptography. Lecture Notes in Computer Science, celebrada a Baltimore (Estats Units d'Amèrica) del 12 al 15 de novembre de 2017., In a sense, a two-party protocol achieves fairness if the output from the computation is obtained simultaneously by both parties. A seminal result by Cleve (STOC 1986) states that fairness is impossible, in general. Surprisingly, Gordon et al. (JACM 2011) showed that there exist interesting functions that are computable with fairness. The two results give rise to a distinction between fair functions and unfair ones. The question of characterizing these functions has been studied in a sequence of works leading to the complete characterization of (symmetric) Boolean functions by Asharov et al. (TCC 2015). In this paper, we design new fully secure protocols for functions that were previously unknown to be fair. To this end, our main technical contribution is a generic construction of a fully secure (fair) protocol, starting with a constant-round protocol satisfying limited security requirements. Our construction introduces new conceptual tools for the analysis of fairness that apply to arbitrary (constant-domain) functions. While the characterization remains open, we believe that our results lay the foundation for a deeper understanding of fairness., V. Daza—Research supported by Project TEC2015-66228-P (MINECO/FEDER, UE). N. Makriyannis—Research supported by ERC starting grant 638121.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Robust facial alignment with internal denoising auto-encoder

  • Aspandi, Decky
  • Martínez, Oriol
  • Sukno, Federico Mateo
  • Binefa i Valls, Xavier
Comunicació presentada a: 16th Conference on Computer and Robot Vision (CRV) celebrat del 29 al 31 de maig de 2019 a Kingston, Canadà., The development of facial alignment models is growing rapidly thanks to the availability of large facial landmarked datasets and powerful deep learning models. However, important challenges still remain for facial alignment models to work on images under extreme conditions, such as severe occlusions or large variations in pose and illumination. Current attempts to overcome this limitation have mainly focused on building robust feature extractors with the assumption that the model will be able to discard the noise and select only the meaningful features. However, such an assumption ignores the importance of understanding the noise that characterizes unconstrained images, which has been shown to benefit computer vision models if used appropriately on the learning strategy. Thus, in this paper we investigate the introduction of specialized modules for noise detection and removal, in combination with our state-of-the-art facial alignment module and show that this leads to improved robustness both to synthesized noise and in-the-wild conditions. The proposed model is built by combining two major subnetworks: internal image denoiser (based on the Auto-Encoder architecture) and facial landmark localiser (based on the inception-resnet architecture). Our results on the 300-W and Menpo datasets show that our model can effectively handle different types of synthetic noise, which also leads to enhanced robustness in real-world unconstrained settings, reaching top state-of-the-art accuracy., This work is partly supported by the Spanish Ministry of Economy and Competitiveness under project grant TIN2017-90124-P, the Ramon y Cajal programme, the Maria de Maeztu Units of Excellence Programme (MDM-2015- 0502) and the donation bahi2018-19 to the CMTech at the UPF.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Exploring customer reviews for music genre classification and evolutionary studies

  • Oramas, Sergio
  • Espinosa-Anke, Luis
  • Lawlor, Aonghus
  • Serra, Xavier
  • Saggion, Horacio
Comunicació presentada a la 17th International Society for Music Information Retrieval Conference (ISMIR 2016), celebrada els dies 7 a 11 d'agost de 2016 a Nova York, EUA., In this paper, we explore a large multimodal dataset of about 65k albums constructed from a combination of Amazon customer reviews, MusicBrainz metadata and AcousticBrainz audio descriptors. Review texts are further enriched with named entity disambiguation along with polarity information derived from an aspect-based sentiment analysis framework. This dataset constitutes the cornerstone of two main contributions: First, we perform experiments on music genre classification, exploring a variety of feature types, including semantic, sentimental and acoustic features. These experiments show that modeling semantic information contributes to outperforming strong bag-of-words baselines. Second, we provide a diachronic study of the criticism of music genres via a quantitative analysis of the polarity associated to musical aspects over time. Our analysis hints at a potential correlation between key cultural and geopolitical events and the language and evolving sentiments found in music reviews., This work was partially funded by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE), by the Keystone COST Action IC1302 and by the Insight Centre for Data Analytics under grant number SFI/12/RC/2289.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Unsupervised classification of planning instances

  • Segovia-Aguas, Javier
  • Jiménez, Sergio
  • Jonsson, Anders, 1973-
Comunicació presentada a: The 27th International Conference on Automated Planning and Scheduling, ICAPS 2017, celebrada a Pittsburgh, Estats Units, del 18 al 23 de juny de 2017, In this paper we introduce a novel approach for unsupervised classification of planning instances based on the recent formalism of planning programs. Our approach is inspired by structured prediction in machine learning, which aims at predicting structured information about a given input rather than a scalar value. In our case, each input is an unlabelled classical planning instance, and the associated structured information is the planning program that solves the instance. We describe a method that takes as input a set of planning instances and outputs a set of planning programs, classifying each instance according to the program that solves it. Our results show that automated planning can be successfully used to solve structured unsupervised classification tasks, and invites further exploration of the connection between automated planning and structured prediction., This work is partially supported by grant TIN2015-67959 and the Maria de Maeztu Units of Excellence Programme MDM-2015-0502, MEC, Spain.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

SemEval-2016 Task 2: Interpretable semantic textual similarity

  • Agirre, Eneko
  • Gonzalez Agirre, Aitor
  • Lopez-Gazpio, Iñigo
  • Maritxalar, Montserrat
  • Rigau Claramunt, German
  • Uria, Larraitz
Comunicació presentada al 10th International Workshop on Semantic Evaluation (SemEval-2016), celebrat els dies 16 i 17 de juny de 2016 a San Diego, Califòrnia., The final goal of Interpretable Semantic Textual Similarity (iSTS) is to build systems that explain which are the differences and commonalities between two sentences. The task adds an explanatory level on top of STS, formalized as an alignment between the chunks in the two input sentences, indicating the relation and similarity score of each alignment. The task provides train and test data on three datasets: news headlines, image captions and student answers. It attracted nine teams, totaling 20 runs. All datasets and the annotation guideline are freely available., This material is based in part upon work supported a MINECO grant to the University of the Basque Country (TUNER project TIN2015-65308-C5-1-R). Aitor Gonzalez Agirre and Inigo Lopez-Gazpio are by doctoral grants from MINECO. The IXA group is funded by the Basque Government (A type Research Group).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A comparison of domain-based word polarity estimation using different word embeddings

  • García-Pablos, Aitor
  • Cuadros Oller, Montse
  • Rigau Claramunt, German
Comunicació presentada a la Tenth International Conference on Language Resources and Evaluation (LREC 2016), celebrada els dies 23 a 28 de maig de 2016 a Portorož, Eslovènia., A key point in Sentiment Analysis is to determine the polarity of the sentiment implied by a certain word or expression. In basic Sentiment Analysis systems this sentiment polarity of the words is accounted and weighted in different ways to provide a degree of positivity/negativity. Currently words are also modelled as continuous dense vectors, known as word embeddings, which seem to encode interesting semantic knowledge. With regard to Sentiment Analysis, word embeddings are used as features to more complex supervised classification systems to obtain sentiment classifiers. In this paper we compare a set of existing sentiment lexicons and sentiment lexicon generation techniques. We also show a simple but effective technique to calculate a word polarity value for each word in a domain using existing continuous word embeddings generation methods. Further, we also show that word embeddings calculated on in-domain corpus capture the polarity better than the ones calculated on general-domain corpus., This work has been supported by Vicomtech-IK4 and partially funded by TUNER project (TIN2015-65308-C5-1-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

SemEval-2018 Task 9: hypernym discovery

  • Camacho-Collados, Jose
  • Delli Bovi, Claudio
  • Espinosa-Anke, Luis
  • Oramas, Sergio
  • Pasini, Tommaso
  • Santus, Enrico
  • Shwartz, Vered
  • Navigli, Roberto
  • Saggion, Horacio
Comunicació presentada al 12th International Workshop on Semantic Evaluation (SemEval-2018), celebrat els dies 5 i 6 de juny de 2018 a Nova Orleans, EUA., This paper describes the SemEval 2018 Shared Task on Hypernym Discovery. We put forward this task as a complementary benchmark for modeling hypernymy, a problem which has traditionally been cast as a binary classification task, taking a pair of candidate words as input. Instead, our reformulated task is defined as follows: given an input term, retrieve (or discover) its suitable hypernyms from a target corpus. We proposed five different subtasks covering three languages (English, Spanish, and Italian), and two specific domains of knowledge in English (Medical and Music). Participants were allowed to compete in any or all of the subtasks. Overall, a total of 11 teams participated, with a total of 39 different systems submitted through all subtasks. Data, results and further information about the task can be found at https://competitions. codalab.org/competitions/17119., The authors gratefully acknowledge the economic support in the construction of the datasets from the Maria de Maeztu-UPF Grant provided to Horacio Saggion, Luis Espinosa-Anke and Sergio Oramas; Google Research through the Google Doctoral Fellowship in Natural Language Processing to Jose Camacho-Collados; and Bar-Ilan University through Vered Shwartz. This work is partially supported by the TUNER project (TIN2015- 65308-C5-5-R, MINECO/FEDER, UE), Spanish Ministry of Economy and Competitiveness.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Random vs structured pilot assignment in cell-free massive MIMO wireless networks

  • Attarifar, Masoud
  • Abbasfar, Aliazam
  • Lozano Solsona, Angel
This paper addresses the performance improvement that structured pilot assignment policies can bring about, relative to a random pilot assignment, in cell-free massive MIMO wireless networks. It is shown that structured policies can deliver a multiple-fold reduction in the pilot overhead required to keep pilot contamination at some acceptably low level. While the implementation of the structured policies considered in the paper might require some degree of centralized control, their performance also captures what distributed nonrandom pilot assignment schemes (e.g., greedy or collision-detecting algorithms) can hope to approach., The work of A. Lozano was supported by Project TEC2015-66228-P (MINECO/FEDER, UE) and by the European Research Council under the H2020 Framework Programme/ERC grant agreement 694974.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Detection of speech events and speaker characteristics through photo-plethysmographic signal neural processing

  • Cámbara Ruiz, Guillermo
  • Luque, Jordi
  • Farrús, Mireia
Comunicació presentada a: 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), celebrat de manera virtual a Barcelona del 4 al 8 de maig de 2020., The use of photoplethysmogram signal (PPG) for heart and sleep monitoring is commonly found nowadays in smart-phones and wrist wearables. Besides common usages, it has been proposed and reported that person information can be extracted from PPG for other uses, like biometry tasks. In this work, we explore several end-to-end convolutional neural network architectures for detection of human’s characteristics such as gender or person identity. In addition, we evaluate whether speech/non-speech events may be inferred from PPG signal, where speech might translate in fluctuations into the pulse signal. The obtained results are promising and clearly show the potential of fully end-to-end topologies for automatic extraction of meaningful biomarkers, even from a noisy signal sampled by a low-cost PPG sensor. The AUCs for best architectures put forward PPG wave as biological discriminant, reaching 79% and 89.0%, respectively for gender and person verification tasks. Furthermore, speech detection experiments reporting AUCs around 69% encourage us for further exploration about the feasibility of PPG for speech processing tasks.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Heatmap-guided balanced deep convolution networks for family classification in the wild

  • Aspandi, Decky
  • Martínez, Oriol
  • Binefa i Valls, Xavier
Comunicació presentada a: Proceedings of the 14th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2019) celebrat del 14 al 18 de maig a Lille, France., Automatic kinship recognition using Computer Vision, which aims to infer the blood relationship between individuals by only comparing their facial features, has started to gain attention recently. The introduction of large kinship datasets, such as Family In The Wild (FIW), has allowed large scale dataset modeling using state of the art deep learning models. Among other kinship recognition tasks, family classification task is lacking any significant progress due to its increasing difficulty in relation to the family member size. Furthermore, most current state of-the-art approaches do not perform any data pre-processing (which try to improve models accuracy) and are trained without a regularizer (which results in models susceptible to overfitting). In this paper, we present the Deep Family Classifier (DFC), a deep learning model for family classification in the wild. We build our model by combining two sub-networks: internal Image Feature Enhancer which operates by removing the image noise and provides an additional facial heatmap layer and Family Class Estimator trained with strong regularizers and a compound loss. We observe progressive improvement in accuracy during the validation phase, with a state of the art results of 16.89% is obtained for the track 2 in the RFIW2019 challenge and 17.08% of familly classification task on FIW dataset., This work is partly supported by the Spanish Ministry of Economy and Competitiveness under project grant TIN2017- 90124-P, the Ramon y Cajal programme, the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and the donation bahi2018-19 to the CMTech at the UPF.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Semantic image inpainting through improved Wasserstein generative adversarial networks

  • Vitoria, Patricia
  • Sintes Marcos, Joan
  • Ballester, Coloma
Comunicació presentada a: VISAPP 2019, celebrada a Praga, República Txeca, del 25 al 27 de febrer de 2019, Image inpainting is the task of filling-in missing regions of a damaged or incomplete image. In this work we tackle this problem not only by using the available visual data but also by incorporating image semantics through the use of generative models. Our contribution is twofold: First, we learn a data latent space by training an improved version of the Wasserstein generative adversarial network, for which we incorporate a new generator and discriminator architecture. Second, the learned semantic information is combined with a new optimization loss for inpainting whose minimization infers the missing content conditioned by the available data. It takes into account powerful contextual and perceptual content inherent in the image itself. The benefits include the ability to recover large regions by accumulating semantic information even it is not fully present in the damaged image. Experiments show that the presented method obtains qualitative and quantitative top-tier results in different experimental situations and also achieves accurate photo-realism comparable to state-of-the-art works., The authors acknowledge partial support by MINECO/FEDER UE project, reference TIN2015-70410-C2-1 and by H2020-MSCA-RISE-2017 project, reference 777826 NoMADS.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A comparison of melody extraction methods based on source-filter modelling

  • Bosch, Juan J.
  • Bittner, Rachel M.
  • Salamon, Justin
  • Gómez Gutiérrez, Emilia, 1975-
This work explores the use of source-filter models for pitch/nsalience estimation and their combination with different/npitch tracking and voicing estimation methods for automatic/nmelody extraction. Source-filter models are used/nto create a mid-level representation of pitch that implicitly/nincorporates timbre information. The spectrogram of/na musical audio signal is modelled as the sum of the leading/nvoice (produced by human voice or pitched musical instruments)/nand accompaniment. The leading voice is then/nmodelled with a Smoothed Instantaneous Mixture Model/n(SIMM) based on a source-filter model. The main advantage/nof such a pitch salience function is that it enhances/nthe leading voice even without explicitly separating it from/nthe rest of the signal. We show that this is beneficial/nfor melody extraction, increasing pitch estimation accuracy/nand reducing octave errors in comparison with simpler/npitch salience functions. The adequate combination with/nvoicing detection techniques based on pitch contour characterisation/nleads to significant improvements over stateof-the-art/nmethods, for both vocal and instrumental music., This work is partially supported by the European Union/nunder the PHENICX project (FP7-ICT-601166) and the/nSpanish Ministry of Economy and Competitiveness under/nCASAS project (TIN2015-70816-R) and Maria de Maeztu/nUnits of Excellence Programme (MDM-2015-0502)
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Purely declarative action descriptions are overrated: classical planning with simulators

  • Francès, Guillem
  • Ramírez Jávega, Miquel
  • Lipovetzky, Nir
  • Geffner, Héctor
Comunicació presentada a la Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI 2017), celebrada els dies 19 a 25 d'agost a Melbourne, Austràlia., Classical planning is concerned with problems where a goal needs to be reached from a known initial state by doing actions with deterministic, known effects. Classical planners, however, deal only with classical problems that can be expressed in declarative planning languages such as STRIPS or PDDL. This prevents their use on problems that are not easy to model declaratively or whose dynamics are given via simulations. Simulators do not provide a declarative representation of actions, but simply return successor states. The question we address in this paper is: can a planner that has access to the structure of states and goals only, approach the performance of planners that also have access to the structure of actions expressed in PDDL? To answer this, we develop domain-independent, black box planning algorithms that completely ignore action structure, and show that they match the performance of state-of-the-art classical planners on the standard planning benchmarks. Effective black box algorithms open up new possibilities for modeling and for expressing control knowledge, which we also illustrate., G. Francès is supported by the M. de Maeztu Programme (MDM-2015-0502) and H. Geffner by grant TIN2015-67959-P, both from MINECO, Spain. M. Ramirez and N. Lipovetzky have been partially funded by the Australian DST Group.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

An adaptable lexical simplification architecture for major ibero-romance languages

  • Ferrés, Daniel
  • Saggion, Horacio
  • Gómez Guinovart, Xavier
Comunicació presentada a: EMNLP 2017, Workshop Building Linguistically Generalizable NLP Systems; celebrat el 8 de setembre de 2017 a Copenhagen, Dinamarca., Lexical Simplification is the task of reducing the lexical complexity of textual documents by replacing difficult words with easier to read (or understand) expressions while preserving the original meaning. The development of robust pipelined multilingual architectures able to adapt to new languages is of paramount importance in lexical simplification. This paper describes and evaluates a modular hybrid linguistic-statistical Lexical Simplifier that deals with the four major Ibero-Romance Languages: Spanish, Portuguese, Catalan, and Galician. The architecture of the system is the same for the four languages addressed, only the language resources used during simplification are language specific., This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R and TIN2015- 65308-C5-1-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

In-camera, photorealistic style transfer for on-set automatic grading

  • Zabaleta, Itziar
  • Bertalmío, Marcelo
Comunicació presentada a la SMPTE 2018 Annual Technical Conference of the Society of Motion Picture and Television Engineers, celebrada a Los Angeles (Estats Units d'Amèrica) del 22 a 25 de octubre de 2018., In professional cinema, the intended artistic look of the movie informs the creation of a static 3D LUT that is applied on set, where further manual modifications to the image appearance are registered as 10-parameter transforms in a color decision list (CDL). The original RAW footage and its corresponding LUT and CDL are passedon to the post-production stage where the fine-tuning of the final look is performed during color grading. — In many cases, the director wants to emulate the style and look present in a reference image, e.g. a still from an existing movie, or a photograph, or a painting, or even a frame from a previously shot sequence in the current movie. The manual creation of a LUT and CDL for this purpose may require a significant amount of work from very skilled artists and technicians, while the state of the art in the academic literature offers promising but partial solutions to the photorealistic style transfer problem, with limitations regarding artifacts, speed and manual interaction. — In this paper, we propose a method that automatically transfers the style, in terms of luminance, color palette and contrast, from a reference image to the source raw footage. It consists of three separable operations: global luminance matching, global color transfer and local contrast matching. As it just takes into account the statistics of source and reference images, no training is required. The total transform is not static but adapts to the changes in the source footage. The computational complexity of the procedure is extremely low and allows for real-time implementation in-camera, for on-set monitoring. While the method is proposed as a substitute for the need to specify a LUT and a CDL, it's compatible with further refinements performed via LUTs, CDLs and grading, both on-set and in post-production. The results are free from artifacts and provide an excellent approximation to the intended look, bringing savings in pre-production, shooting and post-production time., This work has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement number 761544 (project HDR4EU) and under grant agreement number 780470 (project SAUCE), and by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P (MINECO/FEDER,UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Emotion embedding spaces for matching music to stories

  • Won, Minz
  • Salamon, Justin
  • Bryan, Nicholas J
  • Mysore, Gautham J
  • Serra, Xavier
Content creators often use music to enhance their stories, as it can be a powerful tool to convey emotion. In this paper, our goal is to help creators find music to match the emotion of their story. We focus on text-based stories that can be auralized (e.g., books), use multiple sentences as input queries, and automatically retrieve matching music. We formalize this task as a cross-modal text-to-music retrieval problem. Both the music and text domains have existing datasets with emotion labels, but mismatched emotion vocabularies prevent us from using mood or emotion annotations directly for matching. To address this challenge, we propose and investigate several emotion embedding spaces, both manually defined (e.g., valence/arousal) and data-driven (e.g., Word2Vec and metric learning) to bridge this gap. Our experiments show that by leveraging these embedding spaces, we are able to successfully bridge the gap between modalities to facilitate cross modal retrieval. We show that our method can leverage the well established valence-arousal space, but that it can also achieve our goal via data-driven embedding spaces. By leveraging data-driven embeddings, our approach has the potential of being generalized to other retrieval tasks that require broader or completely different vocabularies., This work was funded by the predoctoral grant MDM2015-0502-17-2 from the Spanish Ministry of Economy and Competitiveness linked to the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Supervised distributional hypernym discovery via domain adaptation

  • Espinosa-Anke, Luis
  • Camacho-Collados, Jose
  • Delli Bovi, Claudio
  • Saggion, Horacio
Comunicació presentada a la Conference on Empirical Methods in Natural Language Processing celebrada els dies 1 a 5 de novembre de 2016 a Austin, Texas., Lexical taxonomies are graph-like hierarchical structures that provide a formal representation of knowledge. Most knowledge graphs to date rely on is-a (hypernymic) relations as the backbone of their semantic structure. In this paper, we propose a supervised distributional framework for hypernym discovery which operates at the sense level, enabling large-scale automatic acquisition of disambiguated taxonomies. By exploiting semantic regularities between hyponyms and hypernyms in embeddings spaces, and integrating a domain clustering algorithm, our model becomes sensitive to the target data. We evaluate several configurations of our approach, training with information derived from a manually created knowledge base, along with hypernymic relations obtained from Open Information Extraction systems. The integration of both sources of knowledge yields the best overall results according to both automatic and manual evaluation on ten different domains., This work is partially funded by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and under the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE). The authors also acknowledge support from Dr. Inventor (FP7-ICT-2013.8.1611383). José CamachoCollados is supported by a Google Doctoral Fellowship in Natural Language Processing.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

CobaltF: a fluent metric for MT evaluation

  • Fomicheva, Marina
  • Bel Rafecas, Núria
  • Specia, Lucia
  • da Cunha Fanego, Iria
  • Malinovskiy, Anton
Comunicació presentada a la First Conference on Machine Translation (WMT), que es va dur a terme durant el 54th Annual Meeting of the Association for Computational Linguistics, els dies 7 a 12 d'agost de 2016 a Berlín, Alemanya., The vast majority of Machine Translation (MT) evaluation approaches are based on the idea that the closer the MT output is to a human reference translation, the higher its quality. While translation quality has two important aspects, adequacy and fluency, the existing referencebased metrics are largely focused on the former. In this work we combine our metric UPF-Cobalt, originally presented at the WMT15 Metrics Task, with a number of features intended to capture translation fluency. Experiments show that the integration of fluency-oriented features significantly improves the results, rivalling the best-performing evaluation metrics on the WMT15 data., This work was partially funded by TUNER (TIN2015-65308-C5-5-R) and MINECO/FEDER, UE. Marina Fomicheva was supported by funding from the FI-DGR grant program of the Generalitat de Catalunya. Iria da Cunha was supported by a Ramon y Cajal contract (RYC-2014-16935). Lucia Specia was supported by the QT21 project (H2020 No. 645452).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Latent-based adversarial neural networks for facial affect estimations

  • Aspandi, Decky
  • Mallol Ragolta, Adrià
  • Schuller, Björn
  • Binefa i Valls, Xavier
Comunicació presentada al 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020), celebrat del 16 al 20 de novembre de 2020 a Buenos Aires, Argentina., There is a growing interest in affective computing research nowadays given its crucial role in bridging humans with computers. This progress has recently been accelerated due to the emergence of bigger dataset. One recent advance in this field is the use of adversarial learning to improve model learning through augmented samples. However, the use of latent features, which is feasible through adversarial learning, is not largely explored, yet. This technique may also improve the performance of affective models, as analogously demonstrated in related fields, such as computer vision. To expand this analysis, in this work, we explore the use of latent features through our proposed adversarial-based networks for valence and arousal recognition in the wild. Specifically, our models operate by aggregating several modalities to our discriminator, which is further conditioned to the extracted latent features by the generator. Our experiments on the recently released SEWA dataset suggest the progressive improvements of our results. Finally, we show our competitive results on the Affective Behavior Analysis in-the-Wild (ABAW) challenge dataset., This work is partly supported by the Spanish Ministry of Economy and Competitiveness under project grant TIN2017- 90124-P, the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), and the donation bahi2018-19 to the CMTech at UPF. Further funding has been received from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 826506 (sustAGE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic transcription of flamenco guitar falsetas

  • Rodríguez, Sonia
  • Gómez Gutiérrez, Emilia, 1975-
  • Cuesta, Helena
Comunicació presentada al 8th International Workshop on Folk Music Analysis celebrat del 26 al 29 de juny de 2018 a Tessalònica, Grècia., This work deals with the automatic transcription and characterization of flamenco guitar, with a focus on short melodic interludes improvised between sung verses. These are called falsetas in the flamenco argot and are very challenging for manual and automatic transcription due to their fast and highly ornamented nature. However, they are a key resource for guitar players to practice. We adapted a state of the art singing transcription algorithm to process an audio signal containing one or several guitar falsetas and extract their symbolic representation. The algorithms first perform a segmentation to locate the guitar fragments and then a symbolic transcription of these segments into symbolic representation. In order to evaluate it, we collected the first (to our knowledge) annotated falseta datasets. Our results confirm the difficulty of the task, and a detailed study of two transcriptions revealed that combining the algorithm with specific musical knowledge about the scale used by the song, improves the performance of the system. Our approach follows the principles of research reproducibility, and the system is integrated in a computer-assisted paradigm, where the user complements the automatic annotation with a priory knowledge to generate a final transcription., This work is partially supported by the Spanish Ministry of Economy and Competitiveness under the CASAS project (TIN2015-70816-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

SemEval 2018 Task 2: multilingual emoji prediction

  • Barbieri, Francesco
  • Camacho-Collados, Jose
  • Ronzano, Francesco
  • Espinosa-Anke, Luis
  • Ballesteros, Miguel
  • Basile, Valerio
  • Patti, Viviana
  • Saggion, Horacio
Comunicació presentada al 12th International Workshop on Semantic Evaluation (SemEval-2018), celebrat els dies 5 i 6 de juny de 2018 a Nova Orleans, EUA., This paper describes the results of the first shared task on Multilingual Emoji Prediction, organized as part of SemEval 2018. Given the text of a tweet, the task consists of predicting the most likely emoji to be used along such tweet. Two subtasks were proposed, one for English and one for Spanish, and participants were allowed to submit a system run to one or both subtasks. In total, 49 teams participated in the English subtask and 22 teams submitted a system run to the Spanish subtask. Evaluation was carried out emoji-wise, and the final ranking was based on macro F-Score. Data and further information about this task can be found at https://competitions. codalab.org/competitions/17344., Francesco B. and Horacio S. acknowledge support from the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502). The work of V. Patti and V. Basile was partially funded by the IHatePrejudice project (S1618 L2 BOSC 01).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Trainable citation-enhanced summarization of scientific articles

  • Saggion, Horacio
  • AbuRa'ed, Ahmed Ghassan Tawfiq
  • Ronzano, Francesco
In order to cope with the growing number of relevant scientific publications to consider at a given time, automatic text summarization is a useful technique. However, summarizing scientific papers poses important challenges for the natural language processing community. In recent years a number of evaluation challenges have been proposed to address the problem of summarizing a scientific paper taking advantage of its citation network (i.e., the papers that cite the given paper). Here, we present our trainable technology to address a number of challenges in the context of the 2nd Computational Linguistics Scientific Document/nSummarization Shared Task., This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the European Project Dr. Inventor (FP7-ICT-2013.8.1 - Grant no: 611383).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Solving multiagent planning problems with concurrent conditional effects

  • Furelos Blanco, Daniel
  • Jonsson, Anders, 1973-
Comunicació presentada al 33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Innovative Applications of Artificial Intelligence Conference, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2020, celebrat del 27 de gener a l'1 de febrer de 2019 a Palo Alta, EEUU., In this work we present a novel approach to solving concurrent multiagent planning problems in which several agents act in parallel. Our approach relies on a compilation from concurrent multiagent planning to classical planning, allowing us to use an off-the-shelf classical planner to solve the original multiagent problem. The solution can be directly interpreted as a concurrent plan that satisfies a given set of concurrency constraints, while avoiding the exponential blowup associated with concurrent actions. Our planner is the first to handle action effects that are conditional on what other agents are doing. Theoretically, we show that the compilation is sound and complete. Empirically, we show that our compilation can solve challenging multiagent planning problems that require concurrent actions., This work has been supported by the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502). Anders Jonsson is partially supported by the grants TIN2015-67959 and PCIN-2017-082 of the Spanish Ministry of Science.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic generation of high-level state features for generalized planning

  • Lotinac, Damir
  • Segovia-Aguas, Javier
  • Jiménez, Sergio
  • Jonsson, Anders, 1973-
In many domains generalized plans can only/nbe computed if certain high-level state features,/ni.e. features that capture key concepts to accurately/ndistinguish between states and make good decisions,/nare available. In most applications of generalized/nplanning such features are hand-coded by/nan expert. This paper presents a novel method/nto automatically generate high-level state features/nfor solving a generalized planning problem. Our/nmethod extends a compilation of generalized planning/ninto classical planning and integrates the computation/nof generalized plans with the computation/nof features, in the form of conjunctive queries. Experiments/nshow that we generate features for diverse/ngeneralized planning problems and hence,/ncompute generalized plans without providing a/nprior high-level representation of the states. We/nalso bring a new landscape of challenging benchmarks/nto classical planning since our compilation/nnaturally models classification tasks as classical/nplanning problems., This work is partially supported by grant TIN2015-67959 and the/nMaria de Maeztu Units of Excellence Programme MDM-2015-/n0502, MEC, Spain. Sergio Jimenez is partially supported by the /nJuan de la Cierva program funded by the Spanish government.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Using offline metrics and user behavior analysis to combine multiple systems for music recommendation

  • Ferraro, Andrés
  • Bogdanov, Dmitry
  • Choi, Kyumin
  • Serra, Xavier
Comunicació presentada a: RecSys 2018 Workshop on Offline Evaluation of Recommender Systems, celebrada el 7 d'octubre del 2018 a Vancouver, Canadà., There are many offline metrics that can be used as a reference for evaluation and optimization of the performance of recommender systems. Hybrid recommendation approaches are commonly used to improve some of those metrics by combining different systems. In this work we focus on music recommendation and propose a new way to improve recommendations, with respect to a desired metric of choice, by combining multiple systems for each user individually based on their expected performance. Essentially, our approach consists in predicting an expected error that each system will produce for each user based on their previous activity. To this end, we propose to train regression models for different metrics predicting the performance of each system based on a number of features characterizing previous user behavior in the system. We then use different fusion strategies to combine recommendations generated by each system. Following this approach one can optimize the final hybrid system with respect to the desired metric of choice. As a proof of concept, we conduct experiments combining two recommendation systems, a Matrix Factorization model and a popularity-based recommender. We use the data provided by Melon, a Korean music streaming service, to train and evaluate the performance of the systems., This research has been supported by Kakao Corp., and partially funded by the European Unions Horizon 2020 research and innovation programme under grant agreement No 688382 (AudioCommons) and the Ministry of Economy and Competitiveness of the Spanish Government (Reference: TIN2015-69935-P).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Perceptually-based gamut extension algorithm for emerging wide color gamut display and projection technologies

  • Zamir, Syed Waqas
  • Vazquez-Corral, Javier
  • Bertalmío, Marcelo
Comunicació presentada a: SMPTE 2016 Annual Technical Conference and Exhibition, celebrada a Los Angeles, Estats Units d'Amèrica, del 25 al 27 d'octubre de 2016., While wide color gamut (WCG) capabilities are a key element of emerging display and projection technologies, at present most image content is recorded using standards such as DCI-P3 for cinema or BT.709 for TV that have a reduced color gamut. Therefore, there is a need for gamut extension methods that process regular content and allow to appreciate the full color potential of new displays, improving user experience. We present a gamut extension algorithm that is based on visual perception models and explicitly takes into account low chromatic colors such as skin tones. It produces results that look natural, are free of artifacts of any kind, and outperform the state of the art. The method is fast, allowing for operation interaction if needed., This work was supported by the European Research Council, Starting Grant ref. 306337, by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P (MINECO/FEDER,UE), and by the Icrea Academia Award.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Implications of decentralized Q-learning resource allocation in wireless networks

  • Wilhelmi Roca, Francesc
  • Bellalta, Boris
  • Cano Bastidas, Cristina
  • Jonsson, Anders, 1973-
Comunicació presentada al 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), celebrat els dies 8 a 13 d'octubre de 2017 a Montreal, Canadà., Reinforcement Learning is gaining attention by the wireless networking community due to its potential to learn good-performing configurations only from the observed results. In this work we propose a stateless variation of Q-learning, which we apply to exploit spatial reuse in a wireless network. In particular, we allow networks to modify both their transmission power and the channel used solely based on the experienced throughput. We concentrate in a completely decentralized scenario in which no information about neighbouring nodes is available to the learners. Our results show that although the algorithm is able to find the best-performing actions to enhance aggregate throughput, there is high variability in the throughput experienced by the individual networks. We identify the cause of this variability as the adversarial setting of our setup, in which the most played actions provide intermittent good/poor performance depending on the neighbouring decisions. We also evaluate the effect of the intrinsic learning parameters of the algorithm on this variability., This work has been partially supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), and by the European Regional Development Fund under grant TEC2015-71303-R (MINECO/FEDER).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Monaural score-informed source separation for classical music using convolutional neural networks

  • Miron, Marius
  • Janer Mestres, Jordi
  • Gómez Gutiérrez, Emilia, 1975-
Comunicació presentada a la 18th International Society for Music Information Retrieval Conference (ISMIR 2017), celebrada els dies 23 a 27 d'octubre de 2017 a Suzhou, Xina., Score information has been shown to improve music source separation when included into non-negative matrix factorization (NMF) frameworks. Recently, deep learning approaches have outperformed NMF methods in terms of separation quality and processing time, and there is scope to extend them with score information. In this paper, we propose a score-informed separation system for classical music that is based on deep learning. We propose a method to derive training features from audio files and the corresponding coarsely aligned scores for a set of classical music pieces. Additionally, we introduce a convolutional neural network architecture (CNN) with the goal of estimating time-frequency masks for source separation. Our system is trained with synthetic renditions derived from the original scores and can be used to separate real-life performances based on the same scores, provided a coarse audio-to-score alignment. The proposed system achieves better performance (SDR and SIR) and is less computationally intensive than a score-informed NMF system on a dataset comprising Bach chorales., This work is partially supported by the Spanish Ministry of Economy and Competitiveness under CASAS project (TIN2015-70816-R) and by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Redundancy as a measure of fault-tolerance for the Internet of Things: A review

  • Lobo, Jorge
  • Rullo, Antonino
  • Serra, Edoardo
In this paper we review and analyze redundancy-based fault-tolerant techniques for the IoT as a paradigm to support two of the main goals of computer security: availability and integrity. We organized the presentation in terms of the three main tasks performed by the nodes of an IoT network: sensing, routing, and control. We first discuss how the implementation of fault-tolerance in the three areas is primary for the correct operation of an entire system. We provide an overview of the different approaches that have been used to address failures in sensing and routing. Control devices typically implement state machines that take decisions based on the measurement of sensors and may also ask actuators to execute actions. Traditionally state-machine replication for fault-tolerance is realized through consensus protocols. Most protocols were developed in the 80’s and 90’s. We will review the properties of such protocols in detail and discuss their limitations for the IoT. Since 2008, consensus algorithms took a new direction with the introduction of the concept of blockchain. Standard blockchain based protocols cannot be applied without modifications to support fault-tolerance in the IoT. We will review some recent results in this new class of algorithms, and show how they can provide the flexibility required to support fault-tolerance in control devices, and thus overcome some of the limitations of the traditional consensus protocols., Jorge Lobo was partially supported by the Spanish Ministry of Economy and Competitiveness. Grant Numbers: TIN201681032P, MDM20150502.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Exploring emoji usage and prediction through a temporal variation lens

  • Barbieri, Francesco
  • Marujo, Luís
  • Karuturi, Pradeep
  • Brendel, William
  • Saggion, Horacio
Comunicació presentada a l'International Workshop on Emoji Understanding and Applications in Social Media (Emoji) celebrat el 25 de juny de 2018) a Stanford, Estats Units., The frequent use of Emojis on social media platforms has created a new form of multimodal social interaction. Developing methods for the study and representation of emoji semantics helps to improve future multimodal communication systems. In this paper we explore the usage and semantics of emojis over time. We compare emoji embeddings trained on a corpus of different seasons and show that some emojis are used differently depending on the time of the year. Moreover, we propose a method to take into account the time information for emoji prediction systems, outperforming state-of-the-art systems. We show that, using the time information, the accuracy of some emojis can be significantly improved., Part of this work was done when Francesco B. interned at Snap Inc. Francesco B. and Horacio S. acknowledge support from the TUNER project (TIN2015-65308- C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015- 0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

PDFdigest: an adaptable layout-aware PDF-to-XML textual content extractor for scientific articles

  • Ferrés, Daniel
  • Saggion, Horacio
  • Ronzano, Francesco
  • Bravo Serrano, Àlex, 1984-
Comunicació presentada a la Language Resources and Evaluation Conference (LREC) 2018, celebrada els dies 7 a 12 de maig de 2018 a Miyazaki, Japó., The availability of automated approaches and tools to extract structured textual content from PDF articles is essential to enable scientific text mining. This paper describes and evaluates the PDFdigest tool, a PDF-to-XML textual content extraction system specially designed to extract scientific articles’ headings and logical structure (title, authors, abstract,...) and its textual content. The extractor deals with both text-based and image-based PDF articles using custom rule-based algorithms implemented with existing state-of-the-art open-source tools for both PDF-to-HTML conversion and image-based PDF Optical Character Recognition., This work was partly funded by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Spanish MINECO Ministry (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Monoaural audio source separation using deep convolutional neural networks

  • Chandna, Pritish
  • Miron, Marius
  • Janer Mestres, Jordi
  • Gómez Gutiérrez, Emilia, 1975-
Comunicació presentada a 13th International Conference on Latent Variable Analysis and Signal Separation, celebrada a Grenoble (França) els dies 21 a 23 de febrer de 2017., In this paper we introduce a low-latency monaural source separation framework using a Convolutional Neural Network (CNN). We use a CNN to estimate time-frequency soft masks which are applied for source separation. We evaluate the performance of the neural network on a database comprising of musical mixtures of three instruments: voice, drums, bass as well as other instruments which vary from song to song. The proposed architecture is compared to a Multilayer Perceptron (MLP), achieving on-par results and a significant improvement in processing time. The algorithm was submitted to source separation evaluation campaigns to test efficiency, and achieved competitive results., This work is partially supported by the Spanish Ministry of Economy and Competitiveness under CASAS project (TIN2015-70816-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic playlist continuation using a hybrid recommender system combining features from text and audio

  • Ferraro, Andrés
  • Bogdanov, Dmitry
  • Yoon, Jisang
  • Kim, KwangSeob
  • Serra, Xavier
Comunicació presentada a: RecSys Challenge 2018, celebrat el 7 d'octubre de 2018, a Vancouver, Canadà., The ACM RecSys Challenge 2018 focuses on music recommendation in the context of automatic playlist continuation. In this paper, we describe our approach to the problem and the final hybrid system that was submitted to the challenge by our team Cocoplaya. This system consists in combining the recommendations produced by two different models using ranking fusion. The first model is based on Matrix Factorization and it incorporates information from tracks’ audio and playlist titles. The second model generates recommendations based on typical track co-occurrences considering their proximity in the playlists. The proposed approach is efficient and achieves a good overall performance, with our model ranked 4th on the creative track of the challenge leaderboard., This research has been supported by Kakao Corp., and partially funded by the European Unions Horizon 2020 research and innovation programme under grant agreement No 688382 (AudioCommons) and the Ministry of Economy and Competitiveness of the Spanish Government (Reference: TIN2015-69935-P).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Performance evaluation of objective quality metrics on HLG-based HDR image coding

  • Sugito, Yasuko
  • Bertalmío, Marcelo
Comunicació presentada a: 2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP), celebrada a Anaheim, Estats Units d'Amèrica, del 26 al 29 de novembre de 2018., We evaluate the performance of objective quality metrics for high dynamic range (HDR) image coding that uses the transfer function (TF) of the Hybrid Log-Gamma (HLG) method. Previous evaluations of objective metrics for HDR image coding have studied which of them are reliable predictors of perceived quality; however, in those tests, all the non-linear transforms used both for encoding and by the best-performing metrics are essentially very similar and based on visual perception data of detection thresholds for lightness variations. The HLG non-linearity on the other hand is very different, as it is designed for backward compatibility with standard dynamic range (SDR) displays. We test a variety of options for objective metrics, including HLG-based. Our results show that the ranking of metrics for HDR coding changes drastically depending on the TF used for compression., This work has partially received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement number 761544 (project HDR4EU) and under grant agreement number 780470 (project SAUCE), and by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P(MINECO/FEDER, UE)
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Multimodal metric learning for tag-based music retrieval

  • Won, Minz
  • Oramas, Sergio
  • Nieto Caballero, Oriol
  • Gouyon, Fabien
  • Serra, Xavier
Comunicació presentada a 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2021), celebrat del 6 a l'11 de juny de 2021 de manera virtual., Tag-based music retrieval is crucial to browse large-scale mu-sic libraries efficiently. Hence, automatic music tagging has been actively explored, mostly as a classification task, which has an inherent limitation: a fixed vocabulary. On the other hand, metric learning enables flexible vocabularies by using pretrained word embeddings as side information. Also, met-ric learning has proven its suitability for cross-modal retrieval tasks in other domains (e.g., text-to-image) by jointly learning a multimodal embedding space. In this paper, we investigate three ideas to successfully introduce multimodal metric learning for tag-based music retrieval: elaborate triplet sampling, acoustic and cultural music information, and domain-specific word embeddings. Our experimental results show that the proposed ideas enhance the retrieval system quantitatively and qualitatively. Furthermore, we release the MSD500: a subset of the Million Song Dataset (MSD) containing 500 cleaned tags, 7 manually annotated tag categories, and user taste profiles., This work was funded by the predoctoral grant MDM-2015-0502-17-2 from the Spanish Ministry of Economy and Competitiveness linked to the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Towards large scale multimedia indexing: a case study on person discovery in broadcast news

  • Le, Nam
  • Bredin, Hervé
  • Sargent, Gabriel
  • India, Miquel
  • López Otero, Paula
  • Barras, Claude
  • Guinaudeau, Camille
  • Gravier, Guillaume
  • Barbosa da Fonseca, Gabriel
  • Lyon Freire, Izabela
  • Patrocínio, Zenilton
  • Guimarães, Silvio Jamil F.
  • Martí Juan, Gerard
  • Morros, Josep Ramon
  • Hernando, Javier
  • Docio Fernández, Laura
  • García Mateo, Carmen
  • Meignier, Sylvain
  • Odobez, Jean-Marc
Comunicació presentada a: the 15th International Workshop on Content-Based Multimedia Indexing (CBMI'17), celebrat a Florència, Itàlia, del 19 al 21 de juny de 2017, The rapid growth of multimedia databases and the human interest in their peers make indices representing the location and identity of people in audio-visual documents essential for searching archives. Person discovery in the absence of prior identity knowledge requires accurate association of audio-visual cues and detected names. To this end, we present 3 different strategies to approach this problem: clustering-based naming, verification-based naming, and graph-based naming. Each of these strategies utilizes different recent advances in unsupervised face / speech representation, verification, and optimization. To have a better understanding of the approaches, this paper also provides a quantitative and qualitative comparative study of these approaches using the associated corpus of the Person Discovery challenge at MediaEval 2016. From the results of our experiments, we can observe the pros and cons of each approach, thus paving the way for future promising research directions., This work was supported by the EU project EUMSSI (FP7-611057), ANR project MetaDaTV (ANR-14-CE24-0024) project, Camomile project (PCIN-2013-067), and the projects TEC2013-43935-R, TEC2015-69266-P, TEC2016-75976-R, TEC2015-65345-P financed by the Spanish government and ERDF.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A multilingual predicate matrix

  • Lopez de Lacalle, Maddalen
  • Laparra, Egoitz
  • Aldabe, Itziar
  • Rigau Claramunt, German
Comunicació presentada a la Tenth International Conference on Language Resources and Evaluation (LREC 2016), celebrada els dies 23 a 28 de maig de 2016 a Portorož, Eslovènia., This paper presents the Predicate Matrix 1.3, a lexical resource resulting from the integration of multiple sources of predicate information including FrameNet, VerbNet, PropBank and WordNet. This new version of the Predicate Matrix has been extended to cover nominal predicates by adding mappings to NomBank. Similarly, we have integrated resources in Spanish, Catalan and Basque. As a result, the Predicate Matrix 1.3 provides a multilingual lexicon to allow interoperable semantic analysis in multiple languages., This work has been partially funded by NewsReader (FP7-ICT-2011-8-316404) and TUNER (TIN2015-65308-C5-1-R) projects.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Multilingual CALL framework for automatic language exercise generation from free text

  • Perez Miguel, Naiara
  • Cuadros Oller, Montse
Comunicació presentada a la 15th Conference of the European Chapter of the Association for Computatinal Linguistics (EACL 2017), celebrada els dies 3 a 7 d'abril de 2017 a València, Espanya., This paper describes a web-based application to design and answer exercises for language learning. It is available in Basque, Spanish, English, and French. Based on open-source Natural Language Processing (NLP) technology such as word embedding models and word sense disambiguation, the system enables users to create automatically and in real time three types of exercises, namely, Fill-inthe-Gaps, Multiple Choice, and Shuffled Sentences questionnaires. These are generated from texts of the users’ own choice, so they can train their language skills with content of their particular interest., This work has been funded by the Spanish Government (project TIN2015-65308-C5-1-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A multiclass anisotropic Mumford-Shah functional for segmentation of d-dimensional vectorial images

  • Garamendi, Juan Francisco
  • Schiavi, Emanuele
Comunicació presentada a la 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2017), celebrada del 27 de febrer l'1 de març de 2017 a Porto, Portugal., We present a general model for multi-class segmentation of multi-channel digital images. It is based on the minimization of an anisotropic version of the Mumford-Shah energy functional in the class of piecewise constant functions. In the framework of geometric measure theory we use the concept of common interphases between regions (classes) and the value of the jump discontinuities of the (weak) solution between adjacent regions in order to define a minimal partition energy functional. The resulting problem is non-smooth and non-convex. Non-smoothness is dealt with highlighting the relationship of the proposed model with the well known Rudin, Osher and Fatemi model for image denoising when piecewise constant solutions (i.e partitions) are considered. Non-convexity is tackled with an optimal threshold of the ROF solution which we which generalize to multi-channel images through a probabilistic clustering. The optimal solution is then computed with a fixed point iteration. The result ing algorithm is described and results are presented showing the successful application of the method to Light Field (LF) images., The authors acknowledge partial support by TIN2015-70410-C2-1-R (MINECO/FEDER, UE) and by GRC reference 2014 SGR 1301, Generalitat de Catalunya.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Extending WordNet with fine-grained collocational information via supervised distributional learning

  • Espinosa-Anke, Luis
  • Camacho-Collados, Jose
  • Rodríguez Fernández, Sara
  • Saggion, Horacio
  • Wanner, Leo
Comunicació presentada a COLING 2016: the 26th International Conference on Computational Linguistics; 2016 Dec. 11-16; Osaka (Japan), WordNet is probably the best known lexical resource in Natural Language Processing. While it/nis widely regarded as a high quality repository of concepts and semantic relations, updating and/nextending it manually is costly. One important type of relation which could potentially add enormous/nvalue toWordNet is the inclusion of collocational information, which is paramount in tasks/nsuch as Machine Translation, Natural Language Generation and Second Language Learning. In/nthis paper, we present ColWordNet (CWN), an extended WordNet version with fine-grained collocational/ninformation, automatically introduced thanks to a method exploiting linear relations/nbetween analogous sense-level embeddings spaces. We perform both intrinsic and extrinsic evaluations,/nand release CWN for the use and scrutiny of the community., This work was partially funded by the following projects: MULTISENSOR (FP7-ICT-610411),/nKRISTINA (H2020-645012-RIA), HARENES (FFI201130219-CO2-02), through a predoctoral grant/n(BES-2012-057036), the Mar´ıa de Maeztu Units of Excellence Program (MDM-2015-0502) and TUNER/n(TIN2015-65308-C5-5-R, MINECO/FEDER, UE). Jose Camacho-Collados is supported by a Google/nDoctoral Fellowship in Natural Language Processing.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Are emojis predictable?

  • Barbieri, Francesco
  • Ballesteros, Miguel
  • Saggion, Horacio
Comunicació presentada a la 15th Conference of the European Chapter of the Association for Computational Linguistics, celebrada els dies 3 a 7 d'abril de 2017 a València, Espanya., Emojis are ideograms which are naturally combined with plain text to visually complement or condense the meaning of a message. Despite being widely used in social media, their underlying semantics have received little attention from a Natural Language Processing standpoint. In this paper, we investigate the relation between words and emojis, studying the novel task of predicting which emojis are evoked by text-based tweet messages. We train several models based on Long ShortTerm Memory networks (LSTMs) in this task. Our experimental results show that our neural model outperforms two baselines as well as humans solving the same task, suggesting that computational models are able to better capture the underlying semantics of emojis., First and third authors acknowledge support from the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Melody extraction based on a source-filter model using pitch contour selection

  • Bosch, Juan J.
  • Gómez Gutiérrez, Emilia, 1975-
This work proposes a melody extraction method which combines a pitch salience function based on source-filter modelling with melody tracking based on pitch contour selection. We model the spectrogram of a musical audio signal as the sum of the leading voice and accompaniment. The leading voice is modelled with a Smoothed Instantaneous Mixture Model (SIMM), and the accompaniment is modelled with a Non-negative Matrix Factorization (NMF). The main benefit of this representation is that it incorporates timbre information, and that the leading voice is enhanced, even without an explicit separation from the rest of the signal. Two different salience functions based on SIMM are proposed, in order to adapt the output of such model to the pitch contour based tracking. Candidate melody pitch contours are then created by grouping pitch sequences, using auditory streaming cues. Finally, melody pitch contours are selected using a set of heuristic rules based on contour characteristics and smoothness constraints. The evaluation on a large set of challenging polyphonic music material, shows that the proposed salience functions help increasing the salience of melody pitches in comparison to similar methods. The complete melody extraction methods also achieve a higher overall accuracy than state-of-the-art approaches when evaluated on both vocal and instrumental music., This work is partially supported by the European Union under the PHENICX project (FP7-ICT-601166) and the Spanish Ministry of Economy and Competitiveness under CASAS project (TIN2015-70816-R) and Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Photorealistic style transfer for cinema shoots

  • Zabaleta, Itziar
  • Bertalmío, Marcelo
Comunicació presentada a: 2018 Colour and Visual Computing Symposium (CVCS), celebrat del 19 a 20 de setembre de 2018 a Gjøvik, Noruega, Color grading is the process of subtly mixing andadjusting the color and tonal balance of a movie to achievea specific visuallook. This manual editing task may requirea significant amount of work from very skilled artists andtechnicians. In many cases the director wants to emulate thestyle and look present in a reference image, e.g. a still froman existing movie, a photograph, or even a previously shotsequence in the current movie. In this paper we propose amethod that automatically transfers the style, in terms of tone,color palette and contrast, from a reference image to the sourceRAW image. It consists of three separable operations: globalluminance matching, global color transfer and local contrastmatching. The computational complexity of the procedureis extremely low and allows for real-time implementationin-camera. As it just takes into account the statistics of sourceand reference images, no training is required. The results arefree from artifacts and provide an excellent approximation tothe intended look, bringing savings in pre-production, shootingand post-production time., This work has received funding from the European Union’s Horizon 2020 research and innovation programme undergrant agreement number 761544 (project HDR4EU) and under grant agreement number 780470 (project SAUCE), and by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P (MINECO/FEDER,UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Encouraging children to produce rhythmic beat gestures leads to better narrative discourse performances

  • Prieto Vives, Pilar, 1965-
  • Vilà-Giménez, Ingrid
Recent research has shown that when preschoolers listen to a speaker who is simultaneously making beat gestures, this favors their recall and comprehension of what they have heard and also boosts their narrative performance. However, previous studies have not tested the effect of encouraging children to produce beat gestures while retelling narratives— as opposed to merely observing them—on their narrative performances. In this study, a total of 47 five- and six-year-old children participated in a between-subjects brief training study experiment with a pretest and an immediate posttest design. Children were exposed to a training phase with a total of six one-minute stories, presented under two experimental conditions: 1) beat non-encouraging condition, and 2) beat encouraging condition. Video recordings of the pretest and posttest narratives were then scored for narrative structure and fluency. A comparison of scores showed that children in the group that had been encouraged to use beat gestures performed better than the group of children who were simply asked to retell the story without gesture instruction. All in all, this evidence suggests that encouraging the use of beat gestures by children helps to boost their subsequent narrative performance., This study was approved by the ethics committee of the Universitat Pompeu Fabra, and benefited from funding awarded by the Spanish Ministry of Economy and Competitiveness (FFI2015-66533-P) and by the Generalitat de Catalunya (2014 SGR_925) to the Prosodic Studies Group. The first author also acknowledges an FI grant from the Generalitat de Catalunya (ref. 2017FI_B_00297).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Interpretable emoji prediction via label-wise attention LSTMs

  • Barbieri, Francesco
  • Espinosa-Anke, Luis
  • Camacho-Collados, Jose
  • Schockaert, Steven
  • Saggion, Horacio
Comunicació presentada a la Conference on Empirical Methods in Natural Language Processing, celebrada del 31 d'octubre al 4 de novembre de 2018 a Brussel·les, Bèlgica., Human language has evolved towards newer forms of communication such as social media, where emojis (i.e., ideograms bearing a visual meaning) play a key role. While there is an increasing body of work aimed at the computational modeling of emoji semantics, there is currently little understanding about what makes a computational model represent or predict a given emoji in a certain way. In this paper we propose a label-wise attention mechanism with which we attempt to better understand the nuances underlying emoji prediction. In addition to advantages in terms of interpretability, we show that our proposed architecture improves over standard baselines in emoji prediction, and does particularly well when predicting infrequent emojis., F. Barbieri and H. Saggion acknowledge support from the TUNER project (TIN2015-65308-C5-5- R, MINECO/FEDER, UE). Luis Espinosa-Anke, Jose Camacho-Collados and Steven Schockaert have been supported by ERC Starting Grant 637277.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

How gender and skin tone modifiers affect emoji semantics in Twitter

  • Barbieri, Francesco
  • Camacho-Collados, Jose
Comunicació presentada a la Seventh Joint Conference on Lexical and Computational Semantics, celebrada els dies 5 i 6 de juny de 2018 a Nova Orleans, EUA., In this paper we analyze the use of emojis in social media with respect to gender and skin tone. By gathering a dataset of over twenty two million tweets from United States some findings are clearly highlighted after performing a simple frequency-based analysis. Moreover, we carry out a semantic analysis on the usage of emojis and their modifiers (e.g. gender and skin tone) by embedding all words, emojis and modifiers into the same vector space. Our analyses reveal that some stereotypes related to the skin color and gender seem to be reflected on the use of these modifiers. For example, emojis representing hand gestures are more widely utilized with lighter skin tones, and the usage across skin tones differs significantly. At the same time, the vector corresponding to the male modifier tends to be semantically close to emojis related to business or technology, whereas their female counterparts appear closer to emojis about love or makeup., Francesco B. acknowledges support from the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Multi-task emoji learning

  • Barbieri, Francesco
  • Marujo, Luís
  • Karuturi, Pradeep
  • Brendel, William
Comunicació presentada a la 12th International AAAI Conference on Web and Social Media (ICWSM 2018) celebrada a Stanford (EEUU) el 25 de juny de 2018., Emojis are very common in social media and understanding their underlying semantics is of great interest from a Natural Language Processing point of view. In this work, we investigate emoji prediction in short text messages using a multi-task pipeline that simultaneously predicts emojis, their categories and sub-categories. The categories are either manually predefined in the unicode standard or automatically obtained by clustering over word embeddings. We show that using this categorical information adds meaningful information, thus improving the performance of emoji prediction task. We systematically analyze the performance of the emoji prediction task by varying the number of training samples and also do a qualitative analysis by using attention weights from the prediction task., This work was done when Francesco B. interned at Snap Inc. Francesco B. acknowledge support also from the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

A Framework for multi-f0 modeling in SATB choir recordings

  • Cuesta, Helena
  • Gómez Gutiérrez, Emilia, 1975-
  • Chandna, Pritish
Comunicació presentada a: Sound & Music Computing Conference celebrada del 28 al 31 de maig de 2019 a Màlaga, Espanya., Fundamental frequency (f0) modeling is an important but relatively unexplored aspect of choir singing. Performance evaluation as well as auditory analysis of singing, whether individually or in a choir, often depend on extracting f0 contours for the singing voice. However, due to the large number of singers, singing at a similar frequency range, extracting the exact individual pitch contours from choir recordings is a challenging task. In this paper, we address this task and develop a methodology for modeling pitch contours of SATB choir recordings. A typical SATB choir consists of four parts, each covering a distinct range of pitches and often with multiple singers each. We first evaluate some state-of-the-art multi-f0 estimation systems for the particular case of choirs with a single singer per part, and observe that the pitch of individual singers can be estimated to a relatively high degree of accuracy. We observe, however, that the scenario of multiple singers for each choir part (i.e. unison singing) is far more challenging. In this work we propose a methodology based on combining a multi-f0 estimation methodology based on deep learning followed by a set of traditional DSP techniques to model f0 and its dispersion instead of a single f0 trajectory for each choir part. We present and discuss our observations and test our framework with different singer configurations., This work is partially supported by the European Commission under the TROMPA project (H2020 770376) and by Spanish Ministry of Economy and Competitiveness under the CASAS project (TIN2015-70816-R). First author is supported by FI Predoctoral Grant from AGAUR (Generalitat de Catalunya).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Multimodal emoji prediction

  • Barbieri, Francesco
  • Ballesteros, Miguel
  • Ronzano, Francesco
  • Saggion, Horacio
Emojis are small images that are commonly included in social media text messages. The combination of visual and textual content in the same message builds up a modern way of communication, that automatic systems are not used to deal with. In this paper we extend recent advances in emoji prediction by putting forward a multimodal approach that is able to predict emojis in Instagram posts. Instagram posts are composed of pictures together with texts which sometimes include emojis. We show that these emojis can be predicted by using the text, but also using the picture. Our main finding is that incorporating the two synergistic modalities, in a combined model, improves accuracy in an emoji prediction task. This result demonstrates that these two modalities (text and images) encode different information on the use of emojis and therefore can complement each other., Francesco B. and Horacio S. acknowledge support from the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE) and the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

LaSTUS/TALN @ CLSciSumm-17: cross-document sentence matching and scientific text summarization systems

  • AbuRa'ed, Ahmed Ghassan Tawfiq
  • Chiruzzo, Luis
  • Saggion, Horacio
  • Accuosto, Pablo
  • Bravo Serrano, Àlex, 1984-
Comunicació presentada a: CL-SciSumm 2017, organitzat com a part de The 2nd Joint Workshop on Bibliometric-enhanced Information Retrieval and Natural Language Processing for Digital Libraries (BIRNDL 2017) i conjuntament amb The 40th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2017), In recent years there has been an increasing interest in approaches to scienti c summarization that take advantage of the citations a research paper has received in order to extract its main contributions. In this context, the CL-SciSumm 2017 Shared Task has been proposed to address citation-based information extraction and summarization. In this paper we present several systems to address three of the CL-SciSumm tasks. Notably, unsupervised systems to match citing and cited sentences (Task 1A), a supervised approach to identify the type of information being cited (Task 1B), and a supervised citation-based summarizer (Task 2)., This work is supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

TALN at SemEval-2016 Task 14: semantic taxonomy enrichment via sense-based embeddings

  • Espinosa-Anke, Luis
  • Ronzano, Francesco
  • Saggion, Horacio
This paper describes the participation of the TALN team in SemEval-2016 Task 14: Semantic Taxonomy Enrichment. The purpose of the task is to find the best point of attachment in WordNet for a set of Out of Vocabulary (OOV) terms. These may come, to name a few, from domain specific glossaries, slang or typical jargon from Internet forums and chatrooms. Our contribution takes as input an OOV term, its part of speech and its associated definition, and generates a set of WordNet synset candidates derived from modelling the term’s definition as a sense embedding representation. We leverage a BabelNet-based vector space representation, which allows us to map the algorithm’s prediction to WordNet. Our approach is designed to be generic and fitting to any domain, without exploiting, for instance, HTML markup in source web pages. Our system performs above the median of all submitted systems, and rivals in performance a powerful baseline based on extracting the first word of the definition with the same partof-speech as the OOV term., This work was partially funded by Dr. Inventor (FP7-ICT-2013.8.1611383), and by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Hand-clapping to the rhythm of newly learned words improves L2 pronunciation: evidence from Catalan and Chinese learners of French

  • Prieto Vives, Pilar, 1965-
  • Baills, Florence
  • Zhang, Yuan
Previous research has shown that rhythmic training enhances phonological speech processing (e.g., [1, 2, 3, 4]). Yet little is known about whether rhythmic training can also help to improve pronunciation in a second language (but see [5, 6, 7]). This study tests the potential benefits of hand-clapping to the rhythm of newly learned French words for the acquisition of pronunciation patterns by Catalan children and Chinese adolescents. In two between-subjects experiments with a pretest and posttest design, participants either repeated words while clapping the rhythmic structure of words or only repeated the words. The French target words were very similar to their lexical counterparts in Catalan, whereas they differed completely from Chinese. Participants’ oral production before and after training was rated for accentedness by three French native speakers. Results showed a significant improvement for the clapping group among the Catalan participants, whereas only a near significant tendency appeared for the Chinese adolescents. Individual musical abilities did not interact significantly with the results in either experiments but working memory played a significant role in Chinese participants’ pronunciation. The results show that a short phonological training session based on highlighting the rhythmic structure of words seems to help improve pronunciation in a foreign language as long as there is little demand placed on semantic processing., This research was supported by funding from the Spanish Ministry of Economy and Competitiveness (FFI2015-66533-P) and the Generalitat de Catalunya (2014 SGR-925). The first author acknowledges a grant awarded by the Dept. of Translation and Language Sciences at UPF.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Experiments in detection of implicit citations

  • AbuRa'ed, Ahmed Ghassan Tawfiq
  • Chiruzzo, Luis
  • Saggion, Horacio
Comunicació presentada al WOSP 2018: 7th International Workshop on Mining Scientific Publications, celebrat el 7 de maig de 2018 a Miyazaki, Japó., The identification of explicit and implicit citations to a given reference paper is important for numerous scientific text mining activities such as citation purpose identification, scientific opinion mining, and scientific summarization. This paper presents experiments on the identification of implicit citations in scientific papers. As in previous work, and relying on an annotated dataset of explicit and implicit citation sentences, we cast the problem as classification, evaluating several machine learning algorithms trained on a set of task-motivated features. We compare our work with the state of the art on the annotated dataset obtaining improved performance. We also annotate a new dataset which we make publicly available to validate our approach. The results on the new dataset confirm our set of features outperforms previously published research., This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Exemplar-based image inpainting using an affine invariant similarity measure

  • Fedorov, Vadim
  • Arias Martínez, Pablo
  • Facciolo Furlan, Gabriele
  • Ballester, Coloma
Comunicació presentada a 11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2016), celebrat del 27 al 29 de febrer de 2016 a Roma, Itàlia., Patch-based approaches are used in state-of-the-art methods for image inpainting. This paper presents a new method for exemplar-based image inpainting using transformed patches. The transformation is determined for each patch in a fully automatic way from a surrounding texture content. We build upon a recent affine invariant patch similarity measure that performs an appropriate patch comparison by automatically adapting the size and shape of the patches. As a consequence, it intrinsically extends the set of available source patches to copy information from. We incorporate this measure into a variational formulation for inpainting and present a numerical algorithm for optimizing it. We show that our method can be applied to complete a perspectively distorted texture as well as to automatically inpaint one view of a scene using other view of the same scene as a source. We present experimental results both for gray and color images, and a comparison with some exemplar-based image inpainting methods., The first, second and fourth authors acknowledge partial support by the MINECO/FEDER project with reference TIN2015-70410-C2-1-R , the MICINN project with reference MTM2012- 30772, and by GRC reference 2014 SGR 1301, Generalitat de Catalunya. The second and third authors were partly founded by the Centre National dEtudes Spatiales (CNES, MISS Project), BPIFrance and R´egion Ile de France, in the framework of the FUI 18 Plein Phare project, the European Research Council (advanced grant Twelve Labours n246961), the Office of Naval research (ONR grant N00014-14-1-0023), and ANRDGA project ANR-12-ASTR-0035.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Restraining and encouraging the use of hand gestures: effects on speech

  • Prieto Vives, Pilar, 1965-
  • Cravotta, Alice
  • Busà, Maria Grazia
Previous studies have investigated the effects of the inability to make hand gestures on speakers’ fluency; however, the question of whether encouraging speakers to gesture affects their fluency has received little attention. This study investigates the effect of restraining (Experiment 1) and encouraging (Experiment 2) hand gestures on the following correlates of speech: speech discourse length (number of words and discourse length in seconds), disfluencies (filled pauses, self-corrections, repetitions, insertions, interruptions, silent pauses), and acoustic properties (speech rate, measures of intensity and pitch). In two experiments, 10 native speakers of Italian took part in a narration task where they were asked to describe comic strips. Each experiment compared two conditions. In Experiment 1, subjects first received no instructions as to how to behave when narrating. Then they were told to sit on their hands while speaking. In Experiment 2, subjects first received no instructions and were then actively encouraged to use hand gestures. The results showed that restraining gestures leads to quieter and slower paced speech, while encouraging gestures triggers longer speech discourse, faster speech rate and more fluent and louder speech. Thus, both restraining and encouraging hand gestures seem to clearly affect prosodic properties of speech, particularly speech fluency., Authors acknowledge funding from the University of Padua’s Doctoral School of Linguistic, Philological and Literary Sciences, the Spanish Ministry of Science and Innovation (grant FFI2015-66533-P), and a grant awarded by the Generalitat de Catalunya (2014SGR-925) to the Prosodic Studies Group.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Timbre analysis of music audio signals with convolutional neural networks

  • Pons Puig, Jordi
  • Slizovskaia, Olga
  • Gómez Gutiérrez, Emilia, 1975-
  • Serra, Xavier
Comunicació presentada a la EUSIPCO 2017: 25th European Signal Processing Conference, celebrada els dies 28 d'agost a 2 de setembre de 2017 a Kos, Grècia., The focus of this work is to study how to efficiently tailor Convolutional Neural Networks (CNNs) towards learning timbre representations from log-mel magnitude spectrograms. We first review the trends when designing CNN architectures. Through this literature overview we discuss which are the crucial points to consider for efficiently learning timbre representations using CNNs. From this discussion we propose a design strategy meant to capture the relevant time-frequency contexts for learning timbre, which permits using domain knowledge for designing architectures. In addition, one of our main goals is to design efficient CNN architectures – what reduces the risk of these models to over-fit, since CNNs’ number of parameters is minimized. Several architectures based on the design principles we propose are successfully assessed for different research tasks related to timbre: singing voice phoneme classification, musical instrument recognition and music auto-tagging., This work is partially supported by: the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), the CompMusic project (ERC grant agreement 267583) and the CASAS Spanish research project (TIN2015-70816-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Preschoolers use prosodic mitigation strategies to encode polite stance

  • Prieto Vives, Pilar, 1965-
  • Hübscher, Iris
  • Garufi, Martina
While prosody has been shown to act as a syntactic bootstrapper in early language acquisition, little is known about the role that prosody plays in the later development of a child’s ability to communicate pragmatic information such as the expression of politeness. The goal of this paper is to investigate whether preschool children use prosody earlier and more prominently than lexical and morphosyntactic cues to signal a polite stance. To this end, 64 three- to five-year-old Catalan-dominant children participated in a cross-sectional study involving a request production task under four different conditions, with interlocutors either a classmate or an unfamiliar adult (low/high social distance), and the ‘cost’ to the interlocutor’s face either low or high. The results showed that preschool children tend to use mitigating prosodic strategies to encode a polite stance early on and more markedly than they use lexical or morphosyntactic markers. These findings are consistent with what other research has found regarding the prosodic mitigation strategies used by Catalan-speaking adults to mark polite stance., This research was funded by the Spanish Ministry of Science and Innovation (grant FFI2015-66533-P), and a grant awarded by the Generalitat de Catalunya (2014SGR-925) to the Prosodic Studies Group. The first author also acknowledges an FPI grant from the Spanish Ministry of Science and Innovation (BES-2013-065019).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Color-matching shots from different cameras having unknown gamma or logarithmic encoding curves

  • Gil Rodríguez, Raquel
  • Vazquez-Corral, Javier
  • Bertalmío, Marcelo
Comunicació presentada a: SMPTE 2017 Annual Technical Conference and Exhibition, celebrat del 23 al 26 d'octubre de 2017 a Los Angeles, Estats Units d'Amèrica., In cinema and TV it is quite usual to have to work with footage coming from several cameras, which show noticeable color differences among them even if they are all the same model. In TV broadcasts, technicians work in camera control units so as to ensure color consistency when cutting from one camera to another. In cinema post-production, colorists need to manually colormatch images coming from different sources. Aiming to help perform this task automatically, the Academy Color Encoding System (ACES) introduced a color management framework to work within the same color space and be able to use different cameras and displays; however, the ACES pipeline requires to have the cameras characterized previously, and therefore does not allow to work ‘in the wild’, a situation which is very common. We present a color stabilization method that, given two images of the same scene taken by two cameras with unknown settings and unknown internal parameter values, and encoded with unknown non-linear curves (logarithmic or gamma), is able to correct the colors of one of the images making it look as if it was captured with the other camera. Our method is based on treating the in-camera color processing pipeline as a combination of a 3x3 matrix followed by a non-linearity, which allows us to model a color stabilization transformation among two shots as a linear-nonlinear function with several parameters. We find corresponding points between the two images, compute the error (color difference) over them, and determine the transformation parameters that minimize this error, all automatically without any user input. The method is fast and the results have no spurious colors or spatio-temporal artifacts of any kind. It outperforms the state of the art both visually and according to several metrics, and can handle very challenging real-life examples., This work was supported by the European Research Council, Starting Grant ref. 306337, by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P(MINECO/FEDER,UE), and by the Icrea Academia Award. The work of J. Vazquez-Corral was supported by the Spanish government under Grant IJCI-2014-19516.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Generating context-free grammars using classical planning

  • Segovia-Aguas, Javier
  • Jiménez, Sergio
  • Jonsson, Anders, 1973-
Comunicació presentada a: IJCAI International Joint Conference on Artificial Intelligence, celebrada a Melbourne, Austràlia, del 19 al 25 d'agost de 2017., This paper presents a novel approach for generating Context-Free Grammars (CFGs) from small sets of input strings (a single input string in some cases). Our approach is to compile this task into a classical planning problem whose solutions are sequences of actions that build and validate a CFG compliant with the input strings. In addition, we show that our compilation is suitable for implementing the two canonical tasks for CFGs, string production and string recognition., This work is partially supported by grant TIN2015-67959 and the Maria de Maeztu Units of Excellence Programme MDM-2015-0502, MEC, Spain.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

FastLAS: scalable inductive logic programming incorporating domain-specific optimisation criteria

  • Law, Mark
  • Russo, Alessandra
  • Bertino, Elisa
  • Broda, Krysia
  • Lobo, Jorge
Comunicació presentada a The Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20), celebrat del 7 al 12 de febrer de 2020 a Nova York, Estats Units., Inductive Logic Programming (ILP) systems aim to find a set of logical rules, called a hypothesis, that explain a set of examples. In cases where many such hypotheses exist, ILP systems often bias towards shorter solutions, leading to highly general rules being learned. In some application domains like security and access control policies, this bias may not be desirable, as when data is sparse more specific rules that guarantee tighter security should be preferred. This paper presents a new general notion of a scoring function over hypotheses that allows a user to express domain-specific optimisation criteria. This is incorporated into a new ILP system, called FastLAS, that takes as input a learning task and a customised scoring function, and computes an optimal solution with respect to the given scoring function. We evaluate the accuracy of FastLAS over real-world datasets for access control policies and show that varying the scoring function allows a user to target domain-specific performance metrics. We also compare FastLAS to state-of-the-art ILP systems, using the standard ILP bias for shorter solutions, and demonstrate that FastLAS is significantly faster and more scalable., This research was sponsored by the U.S. Army Research Laboratory and the U.K. Ministry of Defence under Agreement Number W911NF-16-3-0001. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Army Research Laboratory, the U.S. Government, the U.K. Ministry of Defence or the U.K. Government. The U.S. and U.K. Governments are authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon. Jorge Lobo was supported by the Spanish Ministry of Economy and Competitiveness under Grant Numbers TIN201681032P, MDM20150502, and the U.S. Army Research Office under agreement number W911NF1910432.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

MultiScien: a bi-lingual natural language processing system for mining and enrichment of scientific collections

  • Saggion, Horacio
  • Ronzano, Francesco
  • Accuosto, Pablo
  • Ferrés, Daniel
Comunicació presentada a: 2nd Joint Workshop on Bibliometric-enhanced Information Retrieval and Natural Language Processing for Digital Libraries (BIRNDL 2017) celebrada a Tokyo l'11 d'agost de 2017., In the current online Open Science context, scientific datasets and tools for deep text analysis, visualization and exploitation play a major role.We present a system for deep analysis and annotation of scientific text collections. We also introduce the first version of the SEPLN Anthology, a bi-lingual (Spanish and English) fully annotated text resource in the field of natural language processing that we created with our system. Moreover, a faceted-search and visualization system to explore the created resource is introduced. All resources created for this paper will be available to the research community., This work is (partly) supported by the Spanish Ministry of Economy and Com- petitiveness under the Maria de Maeztu Units of Excellence Programme (MDM- 2015-0502) and by the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Automatic, fast and perceptually accurate gamut mapping based on vision science models

  • Zamir, Syed Waqas
  • Vazquez-Corral, Javier
  • Bertalmío, Marcelo
Comunicació presentada a: SMPTE 2017 Annual Technical Conference and Exhibition, celebrat del 23 al 26 d'octubre de 2017 a Los Angeles, Estats Units d'Amèrica., Gamut mapping transforms colors of the original (image or video) content to the color palette of the display device with the simultaneous goals of (a) reproducing content accurately while preserving the artistic intent of the original content’s creator and (b) exploiting the full color rendering potential of the target display device. The rapid advancement in display technologies has created a pressing need to develop automatic and fast gamut mapping algorithms that can deal with imagery intended for both conventional and emerging displays. In this paper, we propose a novel framework based on retinal and color perception models from vision science that offers a functionality to perform both gamut reduction and gamut extension, while preserving hue and taking into account the analysis of the colors of the input image. We evaluate the performance of the proposed framework visually and by using a perceptually-based error metric, according to which the gamut-mapped results of our framework outperform those of the state-of-the-art methods., This work was supported by the European Research Council, Starting Grant ref. 306337, by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P (MINECO/FEDER,UE), and by the Icrea Academia Award. The work of Javier Vazquez-Corral was supported by the Spanish government grant IJCI-2014-19516.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

From occlusion to global depth order, a monocular approach

  • Rezaeirowshan, Babak
  • Ballester, Coloma
  • Haro Ortega, Gloria
Comunicació presentada al 11th International Joint Conference (VISIGRAPP), celebrat del 27 al 29 de febrer de 2016 a Roma, Itàlia., Estimating 3D structure of the scene from a single image remains a challenging problem in computer vision. This paper proposes a novel approach to obtain a global depth order of objects by incorporating monocular perceptual cues such as T-junctions and object boundary convexity, which are local indicators of occlusions, together with physical cues, namely ground contact points. The proposed combination of these local cues complement each other and creates a more thorough partial depth order relationship. The different partial orders are then robustly aggregated using a Markov random chain approximation to obtain the most plausible global depth order. Experiments show that the proposed method excels in comparison to state of the art methods., The authors acknowledge partial support by the MINECO/FEDER project with reference TIN2015-70410-C2-1-R, the MICINN project with reference MTM2012-30772, and by GRC reference 2014 SGR 1301, Generalitat de Catalunya.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Making sense of massive amounts of scientific publications: the scientific knowledge miner project

  • Ronzano, Francesco
  • Freire, Ana
  • Sáez-Trumper, Diego
  • Saggion, Horacio
The World Wide Web has become the hugest repository ever for scientific publications and it continues to increase at an unprecedented rate. Nevertheless, this information overload makes the exploration of this content a very time-consuming task. In this landscape, the availability of text mining tools to characterize and explore distinctive features of the scientific literature is mandatory./nWe present the Scientific Knowledge Miner (SKM) Project, that aims to investigate new approaches and frameworks to facilitate the extraction of knowledge from scientific publications across different disciplines. More specifically, we will focus on citation characterization, recommendation and scientific document summarization., This work is supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), by the European Project Dr. Inventor (FP7-ICT-2013.8.1 - Grant: 611383), the Catalonia Trade and Investment Agency (Agència per la competitivitat de l’empresa, ACCIÓ) and the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

The Event and Implied Situation Ontology (ESO): application and evaluation

  • Segers, Roxane
  • Rospocher, Marco
  • Vossen, Piek
  • Laparra, Egoitz
  • Rigau Claramunt, German
  • Minard, Anne-Lyse
Comunicació presentada a la Tenth International Conference on Language Resources and Evaluation (LREC 2016), celebrada els dies 23 a 28 de maig de 2016 a Portorož, Eslovènia., This paper presents the Event and Implied Situation Ontology (ESO), a manually constructed resource which formalizes the pre and post situations of events and the roles of the entities affected by an event. The ontology is built on top of existing resources such as WordNet, SUMO and FrameNet. The ontology is injected to the Predicate Matrix, a resource that integrates predicate and role information from amongst others FrameNet, VerbNet, PropBank, NomBank and WordNet. We illustrate how these resources are used on large document collections to detect information that otherwise would have remained implicit. The ontology is evaluated on two aspects: recall and precision based on a manually annotated corpus and secondly, on the quality of the knowledge inferred by the situation assertions in the ontology. Evaluation results on the quality of the system show that 50% of the events typed and enriched with ESO assertions are correct., The research for this paper has been partially funded by the European Union 7th Framework Programme NewsReader (FP7-ICT-2011-8-316404) and the Spanish national project TUNER (TIN2015-65308-C5-1-R).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Practical use suggests a re-evaluation of HDR objective quality metrics

  • Sugito, Yasuko
  • Bertalmío, Marcelo
Comunicació presentada a: Eleventh International Conference on Quality of Multimedia Experience (QoMEX, 2019), celebrada a Berlín (Alemanya) del 5 al 7 de juny de 2019., Full-reference objective quality metrics for high dynamic range (HDR) images must be validated in terms of their consistency with subjective evaluation results, and previous studies have found a number of metrics that appear to correlate well with the preference of observers. However, those conclusions were based on experiments that do not correspond to the practical use of a current day HDR professional production scenario. In this work, we carry out subjective evaluation experiments for the popular HDR standards of perceptual quantization (PQ) and Hybrid Log-Gamma (HLG), with a state-of-the-art HDR reference monitor used in broadcasting and post-production, and where all the observers participating in the tests are video experts. We find that the ranking of HDR metrics is now substantially different from what was reported earlier, and also that a simple standard dynamic range (SDR) metric can be applied directly to PQ or HLG encoded signals providing excellent results that surpass those of HDR metrics., This work has partially received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement number 761544 (project HDR4EU) and under grant agreement number 780470 (project SAUCE), and by the Spanish government and FEDER Fund, grant ref. TIN2015-71537-P (MINECO/FEDER, UE)
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

An enhanced adversarial network with combined latent features for spatio-temporal facial affect estimation in the wild

  • Aspandi, Decky
  • Sukno, Federico Mateo
  • Schuller, Björn
  • Binefa i Valls, Xavier
Comunicació presentada al 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, celebrat del 8 al 10 de febrer de 2021 a Setúbal, Portugal., Affective Computing has recently attracted the attention of the research community, due to its numerous applications in diverse areas. In this context, the emergence of video-based data allows to enrich the widely used spatial features with the inclusion of temporal information. However, such spatio-temporal modelling often results in very high-dimensional feature spaces and large volumes of data, making training difficult and time consuming. This paper addresses these shortcomings by proposing a novel model that efficiently extracts both spatial and temporal features of the data by means of its enhanced temporal modelling based on latent features. Our proposed model consists of three major networks, coined Generator, Discriminator, and Combiner, which are trained in an adversarial setting combined with curriculum learning to enable our adaptive attention modules. In our experiments, we show the effectiveness of our approach by reporting our competitive results on both the AFEW-VA and SEWA datasets, suggesting that temporal modelling improves the affect estimates both in qualitative and quantitative terms. Furthermore, we find that the inclusion of attention mechanisms leads to the highest accuracy improvements, as its weights seem to correlate well with the appearance of facial movements, both in terms of temporal localisation and intensity. Finally, we observe the sequence length of around 160 ms to be the optimum one for temporal modelling, which is consistent with other relevant findings utilising similar lengths., This work is partly supported by the Spanish Ministry of Economy and Competitiveness under project grant TIN2017-90124-P, the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502), and the donation bahi2018-19 to the CMTech at UPF. Further funding has been received from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 826506 (sustAGE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

ELMDist: a vector space model with words and MusicBrainz entities

  • Espinosa-Anke, Luis
  • Oramas, Sergio
  • Saggion, Horacio
  • Serra, Xavier
Comunicació presentada a: Workshop on Semantic Deep Learning (SemDeep), celebrat amb ESWC 2017, del 28 de maig a l'1 de juny a Portoroz, Eslovènia., Music consumption habits as well as the Music market have changed dramatically due to the increasing popularity of digital audio and streaming services. Today, users are closer than ever to a vast number of songs, albums, artists and bands. However, the challenge remains in how to make sense of all the data available in the Music domain, and how current state of the art in Natural Language Processing and semantic technologies can contribute in Music Information Retrieval areas such as music recommendation, artist similarity or automatic playlist generation. In this paper, we present and evaluate a distributional sense-based embeddings model in the music domain, which can be easily used for these tasks, as well as a device for improving artist or album clus- tering. The model is trained on a disambiguated corpus linked to the MusicBrainz musical Knowledge Base, and following current knowledge-based approaches to sense-level embeddings, entity-related vectors are provided a la WordNet, concatenating the id of the entity and its mention. The model is evaluated both intrinsically and extrinsically in a supervised entity typing task, and released for the use and scrutiny of the community., We would like to thank the anonymous reviewers for their very helpful comments and suggestions for improving the quality of the manuscript. We also acknowledge support from the Spanish Minmistry of Economy and Competitiveness un- der the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) and under the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Cante Flamenco Tech: aprendizaje del cante con la ayuda de la tecnología

  • Gómez Gutiérrez, Emilia, 1975-
  • Guerrero, Alba
  • Rodríguez, Sonia
Comunicació presentada al V Congreso Interdisciplinar Investigación y Flamenco (INFLA) celebrat el 7 i 8 de juny de 2018 a Sevilla., Cante Flamenco Tech es un proyecto pedagógico cuyo objetivo es el de generar materiales docentes online para el aprendizaje del cante flamenco, siguiendo un énfasis tecnolégico. Para ello, el proyecto integra las tecnologías actuales de análisis de sonido y música en su aplicación a uno de los aspectos más característicos del cante flamenco como son sus recursos vocales. Como aspecto innovador en la docencia, se propone un enfoque interdisciplinar que integra los ámbitos de la acústica, el procesado de sonido y el canto. En éste artículo se describen en primer lugar las caracaterísticas vocales que se han seleccionado en el curso. Además de los recursos audiovisuales y de evaluación propios de éste tipo de cursos (víıdeos, cuestionarios y fórums de discusión), nuestra propuesta es la de integrar tecnologías automáticas de análisis de voz cantada para visualizar los aspectos más característicos de la técnica vocal de los cantaores. Después de introducir las estrategias de visualización seleccionada, presentamos los resultados del proyecto y un ejemplo de la utilidad de las visualizaciones para la caracterización expresiva de diferentes interpretaciones de un mismo palo flamenco. Este tipo de curso es novedoso en cuanto al ámbito, orientación tecnológica e impacto potencial en la comunidad de amantes del flamenco y de la tecnología., Cante Flamenco Tech is a pedagogical project intended to create online teaching materials for flamenco singing following a music technology perspective. The project integrates current technologies of audio analysis for describing one of the most important features of flamenco singing: a set of characteristic vocal resources. Our project is interdisciplinary, as it integrates concepts from different disciplines such as acoustics, sound processing and singing practice. The paper first describes the set of vocal resources presented in the course. In addition to audiovisual resources and evaluation strategies commonly found in online teaching material (videos, questionnaires and discussion forums), our project integrates automatic singing analysis tools for visualizing the most important aspects of vocal techniques in singers. After presenting the selected visualization strategies, we provide some results of our analysis and an illustration of the usefulness of visualizations for the expressive characterization of different performances from the same flamenco style. Our project is novel with respect to context, technological orientation and potential impact in the community of flamenco lovers and technologists., Nuestro proyecto cuenta con el apoyo económico del plan de innovación docente PlaCLIK 2017-2018 de la Universidad Pompeu Fabra de Barcelona y del proyecto CASAS, Ministerio de Economía, Industria y Competitividad de España, TIN2015-70816-R.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

On security policy migrations

  • Lobo, Jorge
  • Bertino, Elisa
  • Russo, Alessandra
Comunicació presentada al SACMAT '20: The 25th ACM Symposium on Access Control Models and Technologies, celebrat del 10 al 12 de juny de 2020 a Barcelona, Espanya., There has been over the past decade a rapid change towards computational environments that are comprised of large and diverse sets of devices, many of them mobile, which can connect in flexible and context-dependent ways. Examples range from networks where we can have communications between powerful cloud centers, to the myriad of simple sensor devices on the IoT. As the management of these dynamic environments becomes ever more complex, we want to propose policy migrations as a methodology to simplify the management of security policies by re-utilizing and re-deploying existing policies as the systems change. We are interested in understanding the challenges raised answering the following question: given a security policy that is being enforced in a particular source computational device, what does it entail to migrate this policy to be enforced in a different target device? Because of the differences between devices and because these devices cannot be seen in isolation but in the context where they are deployed, the meaning of the policy enforced in the source device needs to be re-interpreted and implemented in the context of the target device. The aim of the paper is to present a formal framework to evaluate the appropriateness of the migration., This research was sponsored by the U.S. Army Research Laboratory and the U.K. Ministry of Defence under Agreement Number W911NF-16-3-0001. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Army Research Laboratory, the U.S. Government, the U.K. Ministry of Defence or the U.K. Government. The U.S. and U.K. Governments are authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon. Jorge Lobo was also supported by the Spanish Ministry of Economy and Competitiveness under Grant Numbers TIN201681032P, MDM20150502, and the U.S. Army Research Office under agreement number W911NF1910432.
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Able to read my mail: An accessible e-mail client with assistive technology

  • Saggion, Horacio
  • Ferrés, Daniel
  • Sevens, Leen
  • Schuurman, Ineke
  • Ripollés, Marta
  • Rodríguez, Olga
Comunicació presentada a la conferència Web for All Conference on The Future of Accessible Work celebrada del 2 al 4 d'abril de 2017 a Perth, Austràlia., The Able to Include project aims at improving the living conditions of people with intellectual or developmental disabilities (IDD) in key areas of society. One of its focus points concerns improving the integration of people with IDD in the workplace by introducing accessible Web-based tools. This paper describes one of the tools developed as result of the project: an e-mail client with text simplification and other assistive technologies which makes information transmitted over the Internet more understandable to people with IDD therefore facilitating their labor integration. The accessible Web e-mail client has been developed following a User-Centered Design and tested with people with IDD. The results so far are encouraging., This work was partly funded by the ABLE-TO-INCLUDE project (European Commission CIP Grant No. 621055), the TUNER project (TIN2015-65308-C5-5-R, MINECO/FEDER, UE), and the Spanish MINECO Ministry (MDM-2015-0502).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

CATOTRON – A Neural text-to-speech system in catalan

  • Külebi, Baybars
  • Öktem, Alp
  • Peiró Lilja, Àlex
  • Pascual, Santiago
  • Farrús, Mireia
Comunicació presentada a Interspeech 2020 celebrat del 25 al 29 d'octubre de 2020 a Shanghai, Xina., We present Catotron, a neural network-based open-source speech synthesis system in Catalan. Catotron consists of a sequence-to-sequence model trained with two small opensource datasets based on semi-spontaneous and read speech. We demonstrate how a neural TTS can be built for languages with limited resources using found-data optimization and crosslingual transfer learning. We make the datasets, initial models and source code publicly available for both commercial and research purposes., This work was subsidised by the Catalan Department of Culture. A part of the funding comes from the financing administered by the inheritance board of the Generalitat de Catalunya. The last author has been funded by the Agencia Estatal de Investigaci ´on (AEI), Ministerio de Ciencia, Innovaci´on y Universidades and the Fondo Social Europeo (FSE) under grant RYC- 2015-17239 (AEI/FSE, UE). Part of this work has been carried out using an NVIDIA GPU Titan Xp generously provided by NVIDIA Company. This research was also partially supported by the project TEC2015-69266-P (MINECO/FEDER, UE).
Repositorio Digital de la UPF
Publication Contribución de congreso (ConferenceOutput)

Local denoising based on curvature smoothing can visually outperform non-local methods on photographs with actual noise

  • Ghimpeteanu, Gabriela
  • Kane, David
  • Batard, Thomas
  • Levine, Stacey
  • Bertalmío, Marcelo
Comunicació presentada a: IEEE International Conference on Image Processing 2016, celebrada del 25 l 28 de setembre de 2016 a Phoenix, Estats Units d'Amèrica., We propose a fast, local denoising method where the Euclidean curvature of the noisy image is approximated in a regularizing manner and a clean image is reconstructed from this smoothed curvature. User preference tests show that when denoising real photographs with actual noise our method produces results with the same visual quality as the more sophisticated, nonlocal algorithms Non-local Means and BM3D, but at a fraction of their computational cost. These tests also highlight the limitations of objective image quality metrics like PSNR and SSIM, which correlate poorly with user preference., This work was supported by the European Research Council, Starting Grant ref. 306337, by the Spanish government, grant ref. TIN2015-71537-P, and by the Icrea Academia Award.