Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

MSP2P: Multi-Scale Point-based Approach for Optimal Crowd Localization Through Perspective Analysis

  • Redó Nieto, David
  • Aramburu Retegui, Mikel
  • García Castaño, Jorge
  • Sánchez Salmerón, Antonio José|||0000-0003-1896-5356
[EN] Image-based individual localization in densely populated scenes offers practical advantages beyond mere head counting, enabling a broader range of high-level tasks in crowd analysis. Crowd image data contain drastic changes in head sizes caused by the perspective effect. This specific challenge has not been addressed in the literature, as existing localization methods do not consider multi-scale features. To alleviate this issue, we propose a novel Multi-Scale Point-to-Point Network (MSP2P) in which a set of experts are in charge of predicting head locations a at different perspective levels. However, the training procedure requires ground-truth scale information for precise one-to-one matching. For this reason, we develop a simple yet effective method that uses neighbor density information to estimate the scale associated with each head location. Extensive experiments demonstrate that our method outperforms most state-of-the-art methods on relevant counting benchmarks without compromising performance., The work described in this paper is performed in the H2020 project STARLIGHT ( sustainable Autonomy and Resilience for LEAs using AI against High Priority Threats ). This project has received funding from the European Union s Horizon 2020 research and innovation program under grant agreement No 101021797.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Adversarial Machine Learning: Bayesian Perspectives

  • Ríos Insua, David
  • Naveiro, Roi
  • Gallego, Víctor
  • Poulos, Jason
Adversarial Machine Learning (AML) is emerging as a major field aimed at protecting Machine Learning (ML) systems against security threats: in certain scenarios there may be adversaries that actively manipulate input data to fool learning systems. This creates a new class of security vulnerabilities that ML systems may face, and a new desirable property called adversarial robustness essential to trust operations based on ML outputs. Most work in AML is built upon a game-theoretic modeling of the conflict between a learning system and an adversary, ready to manipulate input data. This assumes that each agent knows their opponent’s interests and uncertainty judgments, facilitating inferences based on Nash equilibria. However, such common knowledge assumption is not realistic in the security scenarios typical of AML. After reviewing such game-theoretic approaches, we discuss the benefits that Bayesian perspectives provide when defending ML-based systems. We demonstrate how the Bayesian approach allows us to explicitly model our uncertainty about the opponent’s beliefs and interests, relaxing unrealistic assumptions, and providing more robust inferences. We illustrate this approach in supervised learning settings, and identify relevant future research problems. Supplementary materials for this article are available online., The authors acknowledge the support of the National Science Foundation under Grant DMS-1638521 to the Statistical and Applied Mathematical Science Institute (SAMSI),NC, USA. RN acknowledges the support of CUNEF University. DRI is supported by the AXA-ICMAT Chair and the Spanish Ministry of Science program PID2021-124662OB-I00. VG acknowledges support from grant FPU16-05034. This work is supported by the Severo Ochoa Excellence Programme CEX-2019-000904-S, the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 101021797 (Starlight) and 815003 (Trustonomy), the US NSF grant DMS-1638521 and a grants from the FBBVA (Amalfi), EOARD (FA8655-21-1-7042) and AFOSR (FA-9550-21- 1-0239)., With funding from the Spanish government through the "Severo Ochoa Centre of Excellence" accreditation (CEX2019-000904-S)., Peer reviewed
Archivo Digital UPM
Publication Artículo científico (JournalArticle)

Application of deep reinforcement learning to UAV swarming for ground surveillance

  • Arranz Esteban, Raúl|||0009-0004-3857-543X
  • Carramiñana Jiménez, David|||0000-0002-1943-3098
  • Miguel Vela, Gonzalo C. de|||0000-0003-3454-683X
  • Besada Portas, Juan Alberto|||0000-0002-4330-4050
  • Bernardos Barbolla, Ana M.|||0000-0001-7678-8154
Academica-e. Repositorio Institucional de la Universidad Pública de Navarra
Publication Contribución de congreso (ConferenceOutput)

Temporal analysis of distribution shifts in malware classification for digital forensics

  • Zola, Francesco
  • Bruse, Jan Lukas
  • Galar Idoate, Mikel
In recent years, malware diversity and complexity have increased substantially, so the detection and classification of malware families have become one of the key objectives of information security. Machine learning (ML)-based approaches have been proposed to tackle this problem. However, most of these approaches focus on achieving high classification performance scores in static scenarios, without taking into account a key feature of malware: it is constantly evolving. This leads to ML models being outdated and performing poorly after only a few months, leaving stakeholders exposed to potential security risks. With this work, our aim is to highlight the issues that may arise when applying ML-based classification to malware data. We propose a three-step approach to carry out a forensics exploration of model failures. In particular, in the first step, we evaluate and compare the concept drift generated by models trained using a rolling windows approach for selecting the training dataset. In the second step, we evaluate model drift based on the amount of temporal information used in the training dataset. Finally, we perform an in-depth misclassification and feature analysis to emphasize the interpretation of the results and to highlight drift causes. We conclude that caution is warranted when training ML models for malware analysis, as concept drift and clear performance drops were observed even for models trained on larger datasets. Based on our results, it may be more beneficial to train models on fewer but recent data and re-train them after a few months in order to maintain performance., This work has been partially supported by the European Union’s Horizon 2020 Research and Innovation Programme under the project STARLIGHT (Grant Agreement No. 101021797).