Mila > Team > Pascal Germain

Pascal Germain

Associate Academic Member
Assistant Professor, Université Laval, Canada CIFAR AI Chair

Pascal Germain is an Assistant Professor in the Computer Science and Software Engineering Department at Université Laval. He is a research scientist in machine learning. Until recently, he held his position at Inria, the national research institute for the digital sciences, in France. His main research interests are statistical learning theory, including PAC-Bayesian theory, and learning algorithms.

Publications

2021-12

Learning Stochastic Majority Votes by Minimizing a PAC-Bayes Generalization Bound
Valentina Zantedeschi, Paul Viallard, Emilie Morvant, Rémi Emonet, Amaury Habrard, Pascal Germain and Benjamin Guedj

2021-10

Learning Aggregations of Binary Activated Neural Networks with Probabilities over Representations
Louis Fortier-Dubois, Gaël Letarte, Benjamin Leblanc, François Laviolette and Pascal Germain
arXiv preprint arXiv:2110.15137
(2021-10-28)
dblp.uni-trier.dePDF
A General Framework for the Disintegration of PAC-Bayesian Bounds
Paul Viallard, Pascal Germain, Amaury Habrard and Emilie Morvant
arXiv: Machine Learning
(2021-10-08)
hal.archives-ouvertes.frPDF

2021-09

Self-Bounding Majority Vote Learning Algorithms by the Direct Minimization of a Tight PAC-Bayesian C-Bound
Paul Viallard, Pascal Germain, Amaury Habrard and Emilie Morvant

2021-06

Apprentissage de Vote de Majorité par Minimisation d'une C-Borne PAC-Bayésienne
Paul Viallard, Pascal Germain and Emilie Morvant
CAp 2021
(2021-06-14)
hal.archives-ouvertes.fr
Dérandomisation des Bornes PAC-Bayésiennes
Paul Viallard, Pascal Germain and Emilie Morvant
CAp 2021
(2021-06-14)
hal.archives-ouvertes.fr

2021-02

A General Framework for the Derandomization of PAC-Bayesian Bounds
Paul Viallard, Pascal Germain, Amaury Habrard and Emilie Morvant
arXiv preprint arXiv:2102.08649
(2021-02-16)
ui.adsabs.harvard.edu

2020-10

Implicit Variational Inference: the Parameter and the Predictor Space.
Yann Pequignot, Mathieu Alain, Patrick Dallaire, Alireza Yeganehparast, Pascal Germain, Josée Desharnais and François Laviolette
arXiv preprint arXiv:2010.12995
(2020-10-24)
ui.adsabs.harvard.eduPDF

2020-09

Target to Source Coordinate-wise Adaptation of Pre-trained Models
Luxin Zhang, Pascal Germain, Yacine Kessaci and Christophe Biernacki
ECML 2020
(2020-09-14)
hal.archives-ouvertes.frPDF
Landmark-Based Ensemble Learning with Random Fourier Features and Gradient Boosting
Léo Gautheron, Pascal Germain, Amaury Habrard, Guillaume Metzler, Emilie Morvant, Marc Sebban and Valentina Zantedeschi
ECML 2020
(2020-09-14)
hal.archives-ouvertes.frPDF

2020-08

PAC-Bayesian Contrastive Unsupervised Representation Learning
Kento Nozawa, Pascal Germain and Benjamin Guedj

2020-04

Improved PAC-Bayesian Bounds for Linear Regression
Vera Shalaeva, Alireza Fakhrizadeh Esfahani, Pascal Germain and Mihaly Petreczky

2020-02

PAC-Bayes and Domain Adaptation
Pascal Germain, Amaury Habrard, François Laviolette and Emilie Morvant
Neurocomputing
(2020-02-28)
www.sciencedirect.comPDF

2019-12

Domain Adaptation from a Pre-trained Source Model: Application on fraud detection tasks
Luxin Zhang, Christophe Biernacki, Pascal Germain and Yacine Kessaci
CMStatistics 2019 - 12th International Conference of the ERCIM WG on Computational and Methodological Statistics
(2019-12-14)
hal.archives-ouvertes.fr
Dichotomize and Generalize: PAC-Bayesian Binary Activated Deep Neural Networks
Gaël Letarte, Pascal Germain, Benjamin Guedj and Francois Laviolette
NEURIPS 2019
(2019-12-08)
papers.nips.ccPDF

2019-09

Multiview Boosting by Controlling the Diversity and the Accuracy of View-specific Voters
Anil Goyal, Emilie Morvant, Pascal Germain and Massih-Reza Amini
Neurocomputing
(2019-09-17)
www.sciencedirect.comPDF

Publications collected and formatted using Paperoni