Bound to Disagree: Generalization Bounds via Certifiable Surrogates.
Mathieu Bazinet, Valentina Zantedeschi, Pascal Germain
Oral presentation presented at the Uncertainty In Artificial Intelligence (UAI) conference, Amsterdam (Pays-Bas).
Contributions
Presentations
Machine learning tools to explore the synergistic impact of physicochemical properties of peptides and filtration membranes on peptide migration during electrodialysis with filtration membranes.
Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Jean-Pierre Clément, Pascal Germain, Laurent Bazinet.
Flash presentation at the One Health International Days (OHID) conference, «Integrating Human, Animal, Plant and Environmental Health», Session 4 «Human as part of One-Health», Liège (Belgium).
Machine Learning–Guided Discovery of Nine Novel Antihypertensive and Antidiabetic Peptides Released from Water Lentil Proteins During Gastrointestinal Digestion : A great example of collaborative research in the 2RI SEBioVA Network.
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Oral presentation at the One Health International Days (OHID) conference, session 4 «Human as part of One-Health», Liège (Belgium).
Les lentilles d’eau, source de protéine émergente aux effets santé insoupçonnés, pour l’alimentation de demain ?
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Invited guest. Presented at the Grand Rendez-vous du Savoir, organized by Université Laval, Québec (Qc).
Profil peptidique, bioactivités et identification de 9 nouveaux peptides bioactifs après digestion in vitro de protéines de lentilles d’eau.
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Oral presentation for Bénéfiq Congress 2025 «15 ans à alimenter partenariats, recherches et innovations», session «Complémentarité et diversification des sources protéiques», Québec (Qc).
. L’apprentissage automatique pour la compréhension des bioprocédés et l’identification de peptides antimicrobiens dans le cadre d’une économie circulaire.
Laurent Bazinet, Aurore Cournoyer, Zain Sanchez-Reinoso, Mathieu Bazinet, Juan de Toro-Martín, Marie-Claude Vohl, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss.
Oral presentation at the École d’été Intelligence artificielle et alimentation durable organisée par INAF, FRQNT -Pôle Bioalimentaire-, IID, Université Laval/ SCAI, Sorbonne Université, Paris (France).
Séparation de peptides bioactifs par électrodialyse avec membrane de filtration : Des applications aux coproduits marins à l’intelligence artificielle.
Laurent Bazinet, Rachel Durand, Loïc Hénaux, Aurore Cournoyer, Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Marie-Ève Bernier, Hairati Aboubacar, Juan de Toro-Martín, Jean-Pierre Clément, Marie-Claude Vohl, Rozenn Ravallec, Benoit Cudennec, Geneviève Pilon, Pascal Germain, André Marette.
Presentation at the Tunisian-Canadian Symposium on Marine Products “Challenges, Innovations and Opportunities for Sustainable Management”, Session 3 “Valorization of Marine Co-Products”, Tunis (Tunisia).
Machine learning for bioactive peptide identification and understanding of bio-processes
Laurent Bazinet, Aurore Cournoyer, Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Marie-Ève Bernier, Hairati Aboubacar, Juan de Toro-Martín, Jean-Pierre Clément, Marie-Claude Vohl, Rozenn Ravallec, Benoit Cudennec, Pascal Germain.
Presentation at the GreenFoodTech 2025 international congress “Sustainable transformation for tomorrow’s food”, Saint-Malo (France).
How hydrolysis duration and discoloration impact peptide population and antimicrobial activities of chicken cruor hydrolysates.
Delasa Rahimi, Mathieu Bazinet, Zain Sanchez-Reinoso, Juan de Toro-Martín, Ismail Fliss, Sergey Mikhaylin, Laurent Bazinet.
Presentation at the INAF student symposium “Sustainable Food: Cultivating Our Health and That of Our Planet”, Theme 3 “Peptides and Proteins”, Quebec (QC).
Influence of electric current conditions on peptide migration and fraction antimicrobial activities during electrodialysis with ultrafiltration membrane: a comprehensive machine learning-peptidomic study.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss, Laurent Bazinet
Presentation at the congress OHID- One Health International Days 2024, Session «Environment as part of One Health», Lille (France).
Exploring the modulation of peptide migration and fraction bioactivity by electrical current modes during electrodialysis with ultrafiltration membrane: A comprehensive machine learning-peptidomic study.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss, Laurent Bazinet
Presentation at the 115th AOCS annual meeting and expo 2024 «Beyond chemistry : solving complex problems together» co-located with the Sustainable Protein Forum, Session «protein and co-products», Québec (Canada).
L’apprentissage automatique pour l’optimisation et la compréhension de procédés en industrie bioalimentaire
Laurent Bazinet, Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss
Presentation at the Semaine NumériQc 2024, Québec (Canada).
Apprentissage pseudo-Bayésien d’une représentation nouvelle obtenue grâce aux ondelettes.
Mathieu Bazinet, Pascal Germain
Presented as a part of the "concours de la relève" of the Semaine NumériQc 2023, un rassemblement provincial, Québec (Canada).
Progrès récents en électrodialyse avec membrane de filtration (ÉDMF) pour la séparation de peptides bioactifs : des applications à la mise à l’échelle.
Laurent Bazinet, Aurore Cournoyer, Mathieu Bazinet, Sabita Kadel, Thibaud Geoffroy, Loïc Hénaux
Presented at the provincial conference PROTEO 2023, Québec (Canada).
Pénalisation d'un modèle linéaire généralisé pour l'apprentissage d'un modèle équitable dans le cadre d'assurance.
Mathieu Bazinet, Samuel-Olivier Lépine, Liliane Ouedraogo, Michaël Rioux
Invited presentation for Desjardins Assurance Générale (2023).
Invited presentation for Beneva Assurance (2023).
Presented at the competition Actulab 2022, Québec (Canada).
How to induce selective separation of antimicrobial peptides by electrodialysis with filtration membrane?
Laurent Bazinet, Aurore Cournoyer, Gaétan Daigle, Mathieu Bazinet, Jacinthe Thibodeau
Presented at the international conference ANTIMIC 2022, Hammamet (Tunisia).
Effects of pulsed electric field and polarity reversal on the selectivity of peptides migration during electrodialysis with ultrafiltration membrane.
Aurore Cournoyer, Gaétan Daigle, Mathieu Bazinet, Jacinthe Thibodeau, Laurent Bazinet
Presented at the international congress MELPRO 2022, Prague (Czechia).
Proceedings
Influence of electric current conditions on peptide migration and fraction antimicrobial activities during electrodialysis with ultrafiltration membrane: a comprehensive machine learning-peptidomic study.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss, Laurent Bazinet
Presentation at the congress OHID- One Health International Days 2024, Session «Environment as part of One Health», Lille, France.
Proceeding
How to induce selective separation of antimicrobial peptides by electrodialysis with filtration membrane?
Laurent Bazinet, Aurore Cournoyer, Gaétan Daigle, Mathieu Bazinet, Jacinthe Thibodeau
Proceedings for the international conference ANTIMIC 2022, Hammamet, Tunisie.
Abstract Proceeding
Posters
Bound to Disagree: Generalization Bounds via Certifiable Surrogates.
Mathieu Bazinet, Valentina Zantedeschi, Pascal Germain.
Poster presented at the Uncertainty In Artificial Intelligence (UAI) conference, Amsterdam (Netherlands).
Machine learning tools to explore the synergistic impact of physicochemical properties of peptides and filtration membranes on peptide migration during electrodialysis with filtration membranes.
Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Jean-Pierre Clément, Pascal Germain, Laurent Bazinet.
Poster presented at the One Health International Days (OHID) conference, «Integrating Human, Animal, Plant and Environmental Health», Session 4 «Human as part of One-Health», Liège (Belgium).
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Machine Learning–Guided Discovery of Nine Novel Antihypertensive and Antidiabetic Peptides Released from Water Lentil Proteins During Gastrointestinal Digestion.
Poster presented at the One Health International Days (OHID) conference, session 4 «Human as part of One-Health», Liège (Belgium).
Découverte assistée par apprentissage automatique de 7 nouveaux peptides antihypertenseurs et/ou antidiabétiques issus d’hydrolysats de phytosérum de tofu dans un contexte d’économie circulaire.
Myriam Dumont, Rosie Deschênes Gagnon, Hairati Aboubacar, Mathieu Bazinet, Tristan Muller, Aurore Cournoyer, Benoit Cudennec, Laurent Bazinet.
Poster presented at the Journée de la recherche du département de Sciences des Aliments, Québec (Qc).
Antimicrobial peptides from chicken, calf and turkey cruors. Three times more cruor, three times more antimicrobial peptides?
Mathéo Desmenez, Mathieu Bazinet, Jacinthe Thibodeau, Delasa Rahimi, Houssine Fliss, Véronique Perreault, A.T. Vincent, Laurent Bazinet.
Poster presented at the Journée de la recherche du département de Sciences des Aliments, Québec (Qc).
Découverte assistée par apprentissage automatique de 7 nouveaux peptides antihypertenseurs et/ou antidiabétiques issus d’hydrolysats de phytosérum de tofu dans un contexte d’économie circulaire.
Myriam Dumont, Rosie Deschênes Gagnon, Hairati Aboubacar, Mathieu Bazinet, Tristan Muller, Aurore Cournoyer, Benoit Cudennec, Laurent Bazinet.
Poster presented at the 93rd Congrès de l’Association Canadienne Francophone pour l’Avancement des Sciences (ACFAS), Colloque sur «Économie circulaire et lutte contre les pertes et le gaspillage dans le secteur bioalimentaire : défis et perspectives», Session «Penser circulaire pour une alimentation durable», Trois-Rivières (Qc).
Application of machine learning tools to study the synergistic impact of physicochemical properties of peptides and filtration membranes on peptide migration during electrodialysis with filtration membranes.
Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Jean-Pierre Clément, Pascal Germain, Laurent Bazinet.
Poster presented at the Journée scientifique Protéo-ULaval, Québec (Qc).
Profil peptidique, bioactivités et identification de 9 nouveaux peptides bioactifs après digestion in vitro de protéines de lentilles d’eau.
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Poster presented at the Journée scientifique Protéo-ULaval, Québec (Qc).
Profil peptidique, bioactivités et identification de 9 nouveaux peptides bioactifs après digestion in vitro de protéines de lentilles d’eau.
Tristan Muller, Hairati Aboubacar, Mélissa Tourret, Jacinthe Thibodeau, Mathieu Bazinet, Rozenn Ravallec, Benoit Cudennec, Laurent Bazinet.
Poster presented at the Bénéfiq 2025 congress «15 ans à alimenter partenariats, recherches et innovations», Québec (Qc).
Generalization Bounds via Meta-Learned Model Representations: PAC-Bayes and Sample Compression Hypernetworks.
Benjamin Leblanc, Mathieu Bazinet, Nathaniel D'Amours, Alexandre Drouin, Pascal Germain
Poster presented at the International Conference on Machine Learning (ICML) 2025.
Voir le poster de ICML
Sample Compression Unleashed : New Generalization Bounds for Real Valued Losses.
Mathieu Bazinet, Valentina Zantedeschi, Pascal Germain.
Poster accepted at the international conference on Artificial Intelligence and Statistics 2025.
Poster accepted at the Workshop on Mathematics of Modern Machine Learning of NeurIPS 2024.
Poster accepted at the Workshop on Machine Learning and Compression of NeurIPS 2024.
Poster presented at the Montreal AI symposium (MAIS) 2024, Montréal, Canada.
See the AISTATS poster
See the NeurIPS poster
See the MAIS poster
Application of machine learning tools to study the synergistic impact of physicochemical properties of peptides and filtration membranes on peptide migration during electrodialysis with filtration membranes.
Zain Sanchez-Reinoso, Mathieu Bazinet, Benjamin Leblanc, Jean-Pierre Clément, Pascal Germain, Laurent Bazinet.
Poster accepted at the GreenFoodTech 2025 international congress “Sustainable transformation for tomorrow’s food”.
Sample Compression Hypernetworks: From Generalization Bounds to Meta-Learning.
Benjamin Leblanc, Mathieu Bazinet, Nathaniel D'Amours, Alexandre Drouin, Pascal Germain
Poster presented at the Workshop on Machine Learning and Compression of NeurIPS 2024.
See the NeurIPS poster
Influence of electric current conditions on peptide migration and fraction antimicrobial activities during electrodialysis with ultrafiltration membrane: a comprehensive machine learning-peptidomic study.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Ismail Fliss, Laurent Bazinet
Poster presented at the congress AntiMic 2024, Session «Innovative antimicrobial molecules of animal origin», Lille, France.
How selective peptide migration is induced by electrical current modes during electrodialysis with ultrafiltration: A comprehensive machine learning-based peptidomic approach.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Laurent Bazinet
Poster presented at the Congrès International Bénéfiq 2023, Québec, Canada.
Machine learning-based peptidomic approach to identify electrical current modes associated phenomena inducing selective peptide migration in electrodialysis with ultrafiltration membrane.
Aurore Cournoyer, Mathieu Bazinet, Jean-Pierre Clément, Pier-Luc Plante, Laurent Bazinet
Poster presented at the Symposium International GreenFoodTech 2023, Montréal, Canada.