Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Shared household environment structures the gut microbiota in a large family cohort with central areolar choroidal dystrophy

  • García-Gutiérrez, Enriqueta
  • Ponsero, Alise J.
  • Delgado, Jesús
  • Crispie, Fiona
  • Cotter, Paul D.
  • Cuenca, Nicolás|||/items/9aff7e32-82c7-4c3b-9e61-c644ea7f9ec8
  • Lax, Pedro|||/items/4cc55375-3e6f-40f5-b3dc-8b55eb7dd8a7
Central areolar choroidal dystrophy (CACD) is a rare monogenic hereditary retinal dystrophy characterised by progressive retinal pigment epithelium and choriocapillaris atrophy, photoreceptor loss, oxidative stress, and chronic inflammation. Emerging evidence suggests that the gut microbiota can modulate systemic inflammatory responses and may influence ocular health through the gut-eye axis. However, the contribution of the gut microbiota to CACD remains unexplored. In this pilot study, we investigated the gut microbiota composition of members of a large Spanish family carrying the PRPH2 Arg195Leu mutation associated with CACD. Faecal samples from 33 mutation carriers and 24 non-carrier household members were analysed using whole-metagenome shotgun sequencing. Taxonomic composition and bacterial diversity were compared adjusting for sequencing depth, age, and sex, accounting for shared household environment through paired and mixed-effects analyses. The findings suggest that gut microbiota variation in this CACD cohort is primarily driven by shared environmental factors rather than disease status at this depth of sequencing. Our study highlights the critical importance of accounting for household effects in microbiome studies of familial and rare genetic diseases and provides a robust framework for future investigations into the gut-eye axis using longitudinal or interventional designs.
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Adversarial Robustness of Deep Learning-based Malware Detectors via (De)Randomized Smoothing

  • Gibert Llauradó, Daniel
  • Zizzo, Giulio
  • Le, Quan
  • Planes Cid, Jordi
Deep learning-based malware detectors have been shown to be susceptible to adversarial malware examples, i.e. malware examples that have been deliberately manipulated in order to avoid detection. In light of the vulnerability of deep learning detectors to subtle input file modifications, we propose a practical defense against adversarial malware examples inspired by (de)randomized smoothing. In this work, we reduce the chances of sampling adversarial content injected by malware authors by selecting correlated subsets of bytes, rather than using Gaussian noise to randomize inputs like in the Computer Vision domain. During training, our chunk-based smoothing scheme trains a base classifier to make classifications on a subset of contiguous bytes or chunk of bytes. At test time, a large number of chunks are then classified by a base classifier and the consensus among these classifications is then reported as the final prediction. We propose two strategies to determine the location of the chunks used for classification: (1) randomly selecting the locations of the chunks and (2) selecting contiguous adjacent chunks. To showcase the effectiveness of our approach, we have trained two classifiers with our chunk-based smoothing schemes on the BODMAS dataset. Our findings reveal that the chunk-based smoothing classifiers exhibit greater resilience against adversarial malware examples generated with state-of-the-art evasion attacks, outperforming a non-smoothed classifier and a randomized smoothing-based classifier by a great margin., This project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402 and by MCIN/AEI/10.13039/501100011033/FEDER, UE under the project PID2022-139835NB-C22.
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Enhancing the insertion of NOP instructions to obfuscate malware via deep reinforcement learning

  • Gibert Llauradó, Daniel
  • Fredrikson, Matt
  • Mateu Piñol, Carles
  • Planes Cid, Jordi
Current state-of-the-art research for tackling the problem of malware detection and classification is centered on the design, implementation and deployment of systems powered by machine learning because of its ability to generalize to never-before-seen malware families and polymorphic mutations. However, it has been shown that machine learning models, in partidular deep neural networks, lack robustness against crafted inputs (adversarial examples). In this work, we have investigated the vulnerability of a state-of-the-art shallow convolutional neural network malware classifier against the deat code insertion technique. We propose a general framework powered by a Double Q-network to induce misclassification over malware families. The framework trains an agent through a convolutional neural network to select the optimal positions in a code sequence to insert dead code instructions so that the machine learning classifier mislabels the resulting executable. The experiments show that the proposed method significantly drops the classification accuracy of the classifier to 56.53% while having an evasion rate of 100% for the samples belonging to Kelihos_ver3, Simda, and Kelihos_ver1 families. In addition, the average number of instructions needed to mislabel malware in comparison to a random agent decreased by 33%., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 847402. This research has been partially funded by the Spanish MICINN Projects TIN2015-71799-C2-2-P, ENE2015-64117-C5-1-R, PID2019-111544GB-C22, and supported by the University of Lleida.
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Fusing feature engineering and deep learning: A case study for malware classification

  • Gibert Llauradó, Daniel
  • Planes Cid, Jordi
  • Mateu Piñol, Carles
  • Le, Quan
Machine learning has become an appealing signature-less approach to detect and classify malware because of its ability to generalize to never-before-seen samples and to handle large volumes of data. While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts’ knowledge of the domain, deep learning approaches replace the manual feature engineering process by an underlying system, typically consisting of a neural network with multiple layers, that perform both feature learning and classification altogether. However, the combination of both approaches could substantially enhance detection systems. In this paper we present an hybrid approach to address the task of malware classification by fusing multiple types of features defined by experts and features learned through deep learning from raw data. In particular, our approach relies on deep learning to extract N-gram like features from the assembly language instructions and the bytes of malware, and texture patterns and shapelet-based features from malware’s grayscale image representation and structural entropy, respectively. These deep features are later passed as input to a gradient boosting model that combines the deep features and the hand-crafted features using an early-fusion mechanism. The suitability of our approach has been evaluated on the Microsoft Malware Classification Challenge benchmark and results show that the proposed solution achieves state-of-the-art performance and outperforms gradient boosting and deep learning methods in the literature., This project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402, and from the Spanish Science and Innovation Ministry funded project PID2019-111544GB-C22. 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 CeADAR, University College Dublin, and the University of Lleida.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

A secondary model for the effect of pH on the variability in growth fitness of Listeria innocua strains

  • Garcia-Gutierrez, Enriqueta
  • Monteoliva García, Gonzalo
  • Bodea, Ioana
  • Cotter, Paul D.
  • Iguaz, Asunción
  • Garre, Alberto
6 Pág., Variability in microbial growth is a keystone of modern Quantitative Microbiological Risk Assessment (QMRA). However, there are still significant knowledge gaps on how to model variability, with the most common assumption being that variability is constant. This is implemented by an error term (with constant variance) added on top of the secondary growth model (for the square root of the growth rate). However, this may go against microbial ecology principles, where differences in growth fitness among bacterial strains would be more prominent in the vicinity of the growth limits than at optimal growth conditions. This study coins the term "secondary models for variability", evaluating whether they should be considered in QMRA instead of the constant strain variability hypothesis. For this, 21 strains of Listeria innocua were used as case study, estimating their growth rate by the two-fold dilution method at pH between 5 and 10. Estimates of between-strain variability and experimental uncertainty were obtained for each pH using mixed-effects models, showing the lowest variability at optimal growth conditions, increasing towards the growth limits. Nonetheless, the experimental uncertainty also increased towards the extremes, evidencing the need to analyze both sources of variance independently. A secondary model was thus proposed, relating strain variability and pH conditions. Although the modelling approach certainly has some limitations that would need further experimental validation, it is an important step towards improving the description of variability in QMRA, being the first model of this type in the field., This Research was partly funded with project PID2020-116318RB-C32 by the Spanish Ministry of Science and Research, MCIN/AEI/10.13039/501100011033, as well as by the AGROALNEXT programme and was supported by MCIN with funding from European Union NextGenerationEU (PRTR-C17.I1) and by Fundación Séneca with funding from Comunidad Autónoma Región de Murcia (CARM). Alberto Garre acknowledges being funded by a Maria Zambrano Scholarship and by a Ramon y Cajal Fellowship (RYC-2021-034612-I). This work was supported by Enterprise Ireland and the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement number 847402 awarded to EG-G, who was also funded by a Beatriz Galindo scholarship from the Spanish Ministry of Universities (BG22/00060). Work in the Cotter laboratory is also funded by the European Union's Horizon 2020 research and innovation programme, under the MASTER project (grant number 818368), by Science Foundation Ireland (SFI) under grant number SFI/12/RC/2273_P2 (APC Microbiome Ireland SFI Research Centre), and by SFI together with the Irish Department of Agriculture, Food, and the Marine under grant number SFI/16/RC/3835 (VistaMilk SFI Research Centre). Research in the Cotter laboratory is also funded by Enterprise Ireland and industry in the Food for Health Ireland (FHI)-3 project under grant number TC/2018/0025. Dr. Bodea was supported by an EU-FORA fellowship (GP/EFSA/ENCO/2021/01-GA9)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Assessing the impact of packing on static machine learning-based malware detection and classification systems

  • Gibert, Daniel
  • Totosis, Nikolaos
  • Patsakis, Constantinos
  • Le, Quan
  • Zizzo, Giulio
The proliferation of malware, particularly through the use of packing, presents a significant challenge to static analysis and signature-based malware detection techniques. Applying packing to the original executable code renders extracting meaningful features and signatures challenging. To deal with the increasing amount of malware in the wild, researchers and anti-malware companies started harnessing machine learning capabilities with very promising results. However, little is known about the effects of packing on static machine learning-based malware detection and classification systems. This work addresses this gap by investigating the impact of packing on the performance of static machine learning-based models used for malware detection and classification, with a particular focus on those using visualization techniques. To this end, we present a comprehensive analysis of various packing techniques and their effects on the performance of machine learning-based detectors and classifiers. Our findings highlight the limitations of current static detection and classification systems and underscore the need to be proactive to effectively counteract the evolving tactics of malware authors., This project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402 and was supported by the European Commission under the Horizon Europe Programme, as part of the projects, SafeHorizon (https://www.project-safehorizon.eu/) (Grant Agreement no. 101168562), CyberSecPro (https://www.cybersecpro-project.eu) (Grant Agreement no. 101083594) and LAZARUS (https://lazarus-he.eu/) (Grant Agreement no. 101070303). The content of this article does not reflect the official opinion of the European Union. Responsibility for the information and views expressed therein lies entirely with the authors. We would like to thank Cormac Doherty and UCD’s Centre for Cybersecurity and Cybercrime Investigation for their support., Peer reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Enhancing the insertion of NOP instructions to obfuscate malware via deep reinforcement learning

  • Gibert Llauradó, Daniel
  • Fredrikson, Matt
  • Mateu Piñol, Carles
  • Planes Cid, Jordi
Current state-of-the-art research for tackling the problem of malware detection and classification is centered on the design, implementation and deployment of systems powered by machine learning because of its ability to generalize to never-before-seen malware families and polymorphic mutations. However, it has been shown that machine learning models, in partidular deep neural networks, lack robustness against crafted inputs (adversarial examples). In this work, we have investigated the vulnerability of a state-of-the-art shallow convolutional neural network malware classifier against the deat code insertion technique. We propose a general framework powered by a Double Q-network to induce misclassification over malware families. The framework trains an agent through a convolutional neural network to select the optimal positions in a code sequence to insert dead code instructions so that the machine learning classifier mislabels the resulting executable. The experiments show that the proposed method significantly drops the classification accuracy of the classifier to 56.53% while having an evasion rate of 100% for the samples belonging to Kelihos_ver3, Simda, and Kelihos_ver1 families. In addition, the average number of instructions needed to mislabel malware in comparison to a random agent decreased by 33%., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 847402. This research has been partially funded by the Spanish MICINN Projects TIN2015-71799-C2-2-P, ENE2015-64117-C5-1-R, PID2019-111544GB-C22, and supported by the University of Lleida.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Fusing feature engineering and deep learning: A case study for malware classification

  • Gibert Llauradó, Daniel
  • Planes Cid, Jordi
  • Mateu Piñol, Carles
  • Le, Quan
Machine learning has become an appealing signature-less approach to detect and classify malware because of its ability to generalize to never-before-seen samples and to handle large volumes of data. While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts’ knowledge of the domain, deep learning approaches replace the manual feature engineering process by an underlying system, typically consisting of a neural network with multiple layers, that perform both feature learning and classification altogether. However, the combination of both approaches could substantially enhance detection systems. In this paper we present an hybrid approach to address the task of malware classification by fusing multiple types of features defined by experts and features learned through deep learning from raw data. In particular, our approach relies on deep learning to extract N-gram like features from the assembly language instructions and the bytes of malware, and texture patterns and shapelet-based features from malware’s grayscale image representation and structural entropy, respectively. These deep features are later passed as input to a gradient boosting model that combines the deep features and the hand-crafted features using an early-fusion mechanism. The suitability of our approach has been evaluated on the Microsoft Malware Classification Challenge benchmark and results show that the proposed solution achieves state-of-the-art performance and outperforms gradient boosting and deep learning methods in the literature., This project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402, and from the Spanish Science and Innovation Ministry funded project PID2019-111544GB-C22. 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 CeADAR, University College Dublin, and the University of Lleida.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Adversarial Robustness of Deep Learning-based Malware Detectors via (De)Randomized Smoothing

  • Gibert Llauradó, Daniel
  • Zizzo, Giulio
  • Le, Quan
  • Planes Cid, Jordi
Deep learning-based malware detectors have been shown to be susceptible to adversarial malware examples, i.e. malware examples that have been deliberately manipulated in order to avoid detection. In light of the vulnerability of deep learning detectors to subtle input file modifications, we propose a practical defense against adversarial malware examples inspired by (de)randomized smoothing. In this work, we reduce the chances of sampling adversarial content injected by malware authors by selecting correlated subsets of bytes, rather than using Gaussian noise to randomize inputs like in the Computer Vision domain. During training, our chunk-based smoothing scheme trains a base classifier to make classifications on a subset of contiguous bytes or chunk of bytes. At test time, a large number of chunks are then classified by a base classifier and the consensus among these classifications is then reported as the final prediction. We propose two strategies to determine the location of the chunks used for classification: (1) randomly selecting the locations of the chunks and (2) selecting contiguous adjacent chunks. To showcase the effectiveness of our approach, we have trained two classifiers with our chunk-based smoothing schemes on the BODMAS dataset. Our findings reveal that the chunk-based smoothing classifiers exhibit greater resilience against adversarial malware examples generated with state-of-the-art evasion attacks, outperforming a non-smoothed classifier and a randomized smoothing-based classifier by a great margin., This project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402 and by MCIN/AEI/10.13039/501100011033/FEDER, UE under the project PID2022-139835NB-C22.