Expertises
Cybersécurité , Sciences de l’information
- Professeur
- Directeur Scientifique du Laboratoire de cybersécurité et d’intégrité de l’information
- Membre de Unité mixte de recherche INRS-UQO sur la cybersécurité
Courriel :
anderson.avila@inrs.ca
Téléphone :
+1 (819) 595 3900 – poste 3673
Intérêts de recherche
Apprentissage fédéré pour la confidentialité des données
Décentralisation des modèles d’intelligence artificielle en permettant une formation et un raisonnement efficaces sur les dispositifs d’accès afin de favoriser la confidentialité des données
Cyberdéfense et traitement automatique des langues
Lutte contre la mésinformation à partir d’une série de signaux sémantiques, notamment la parole, le texte et l’image
Biométrie
Amélioration de l’authentification en utilisant les traits physiques et comportementaux humains
Anderson Avila est professeur adjoint du Centre Énergie Matériaux Télécommunications de l’Institut national de la recherche scientifique (INRS) et travaille dans l’Unité mixte de recherche INRS-UQO sur la cybersécurité. Ses recherches portent sur l’apprentissage automatique et le traitement de signal appliqué au traitement automatique des langues. Pendant son doctorat en philosophie, il a travaillé sur l’élaboration de nouveaux modèles pour l’évaluation de la qualité de la parole, et sur la robustesse de la biométrie vocale.
Avant de se joindre à l’INRS-UQO, il était un chercheur scientifique dans le domaine du traitement du langage naturel et de la parole, travaillant sur des projets liés à la compression de modèles, à la faible latence et à la robustesse de la compréhension du langage parlé.
Le Professeur Avila a obtenu son baccalauréat ès sciences en informatique de l’Université fédérale de São Carlos, sa maîtrise ès sciences de l’Université fédérale de l’ABC et son doctorat en télécommunications de l’INRS.
Publications
Wahab, Omar A and Avila, Anderson R. A Max-Min Security Game for Coordinated Backdoor Attacks on Federated Learning.” First International Workshop on Machine Learning for securing IoT systems using BigData (2023).
Raphaël Khoury, Anderson Avila, Jacob Brunelle, Baba Mamadou Camara, “How Secure is Code Generated by ChatGPT?”, IEEE Systems, Man, and Cybernetics (SMC), Maui, HA, USA, Oct. 2023.
H. Guimarães, Y. Zhu, O. Mengara, A. Avila, and T. Falk, Assessing the Vulnerability of Self-Supervised Speech Representations for Keyword Spotting under White-Box Adversarial Attacks, IEEE SMC 2023.
Guimarães, Heitor R., et al. « RobustDistiller: Compressing Universal Speech Representations for Enhanced Environment Robustness. » ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2023, DOI: 10.1109/ ICASSP49357.2023.10095480.
Avila, A.R., Bibi, K., Yang, R.H., Li, X., Xing, C., Chen, X. (2022) Low-bit Shift Network for End-to-End Spoken Language Understanding. Proc. Interspeech 2022, 2698-2702, DOI: 10.21437/Interspeech.2022-760.
Avila, D. O’Shaughnessy, T. Falk, Automatic Speaker Verification from Affective Speech Using Gaussian Mixture Model Based Estimation of Neutral Speech Characteristics, J. Speech Communication, vol. 132, p. 21-31, 2021.
A. Avila, J. Alam, F. Prado, D. O’Shaughnessy, T. Falk, On the Use of Blind Channel Response Estimation and a Residual Neural Network to Detect Physical Access Attacks to Speaker Verification Systems, J. Computer Speech & Language, vol. 66, 2021.
N. Potdar, A. Avila, C. XING, et al. A Streaming End-to-End Framework For Spoken Language Understanding. Proc. Thirtieth International Joint Conference on Artificial Intelligence (IJCAI 2021), DOI: 10.24963/ijcai.2021/538.
Cao, Y., Potdar, N., Avila, A.R. (2021) Sequential End-to-End Intent and Slot Label Classification and Localization. Proc. Interspeech 2021, 1229-1233, DOI: 10.21437/ Interspeech.2021-1569.
A. Avila, D. O’Shaughnessy, T. Falk, Non-intrusive Speech Quality Prediction Based on the Blind Estimation of Clean Speech and the i-vector Framework, J. Quality and User Experience, vol.5, no 1, p. 1-15, 2020.
A. Avila, J. Alam, D. O’Shaughnessy, T. Falk, On the Use of the I-vector Speech Representation for Instrumental Quality Measurement, J. Quality and User Experience, 2020, vol. 5, no 1, pp. 1-14, DOI: https://doi.org/10.1007/s41233-020-00036-z.
A. Avila, J. Alam, D. O’Shaughnessy, T. Falk, Blind Channel Response Estimation for Replay Attack Detection, Interspeech 2019, pp. 2893-2897, DOI: 10.21437/Interspeech.2019-2956.
B. Sadou, A. Lahoulou, T. Bouden, A. Avila, T. Falk, Z. Akhtar, Free-Reference Image Quality Assessment Framework using Metrics Fusion and Dimensionality Reduction, Signal & Image Processing, 2019, Vol. 10, No. 5, pp. 1-14, DOI: 10.5121/sipij.2019.10501.
A. Avila, H. Gamper, C. Reddy, R. Cutler, I. Tashev, J. Gehrke, Non-intrusive speech quality assessment using neural networks, ICASSP 2019 – 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 631-635, DOI: 10.1109/ICASSP.2019.8683175.
A. Avila, S. Kshirsagar, A. Tiwari, D. Lafond, D. O’Shaughnessy, and T. Falk, Speech-Based Stress and Emotion Classification Based on Modulation Spectral Features and Convolutional Neural Networks, 27th European Signal Processing Conference (EUSIPCO) 2019, pp. 1-5, DOI: 10.23919/EUSIPCO.2019.8903014.
B. Sadou, A. Lahoulou, T. Bouden, A. Avila, T. Falk, Z. Akhtar, Blind Image Quality Assessment using SVD based Dominant Eigenvectors for Feature Selection, SIPRO 2019.
A. Avila, J. Alam, D. O’Shaughnessy, T. Falk, Intrusive Quality Measurement of Noisy and Enhanced Speech based on i-Vector Similarity, QoMEX 2019, pp. 1-5, DOI: 10.1109/QoMEX.2019.8743285.
A. Avila, Z. Akhtar, J. Santos, D. O’Shaughnessy, T. Falk, Feature Pooling for Spontaneous Speech-Based Emotion Recognition in-the-wild, IEEE Transaction on Affective Computing, 2018, pp. 1-12, DOI: 10.1109/TAFFC.2018.2858255.
A. Avila, J. Alam, D. O’Shaughnessy, T. Falk, Investigating Speech Enhancement and Perceptual Quality for Speech Emotion Recognition, Interspeech 2018, pp. 3663-3667, DOI: 10.21437/Interspeech.2018-2350..
A. Avila, J. Monteiro, D. O’Shaughnessy, T. Falk, Speech Emotion Recognition on Mobile Devices Using a New Modulation Spectrum Pooling and Deep Neural Networks, ISSPIT 2017, 2017 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), pp. 360-365, DOI: 10.1109/ISSPIT.2017.8388669.
A. Avila, B. Cauchi, S. Goetze, S. Doclo, T. Falk, Performance Comparison of Intrusive and Non-instrusive Instrumental Quality Measures for Enhanced Speech, IWAENC 2016, 2016 IEEE International Workshop on Acoustic Signal Enhancement (IWAENC), pp. 1-5, DOI: 10.1109/IWAENC.2016.7602907.
A. Avila, M. Santos, F. Fraga, and T. Falk, The Effect of Speech Rate on Automatic Speaker Verification: a Comparative Analysis of GMM-UBM and I-vector Based Methods, 12th Audio Engineering Conference (AES-Brazil), May 2014.
Avila, M. Paja, F. Fraga, D. O’Shaughnessy, and T. Falk, Improving the Performance of Far-Field Speaker Verification Using Multi-Condition Training: The Case of GMM-UBM and i-vector Systems, Interspeech’2014.
A. Avila, M. Santos, F. Fraga, and T. Falk, Investigating the use of Modulation Spectral Features within an Ivector Framework for Far-Field Automatic Speaker Verification, International. Telecommunications Symposium, 2014.
A. Avila, F. Prado, G. Kobayashi, E. Rocha, Performance Comparison of Overdetermined Multilateration Algorithms for Estimating Aircraft Position. In: Workshop on Distance Geometry and Applications (DGA), 2013, Manaus.
A. Avila, M. Paja, F. Fraga, Proposta de um Sistema de Diálogo Automático Baseado em Algoritmos de Aprendizado Por Reforço. In: Proceedings of the 10th AES Brazil Conference. Rio de Janeiro: Audio Engineering Society, 2012. v. 1. p. 75-78.
A. Avila, M. Paja, F. Fraga, Integracão de Sistemas de Reconhecimento, Tradução e Síntese Automática da Fala para Facilitar a Comunicação de Turistas. In: The 14th LAC AES Conference. Montevideo: Audio Engineering Society, 2011. v. 1, p. 1-4.