Adam Oberman

Membre Académique Associé
Adam Oberman
Professeur, McGill University
Adam Oberman

Adam Oberman is a Professor in the Department of Mathematics and Statistics at McGill University, and director of the Applied Mathematics Laboratory at the Centre de Recherches Mathematiques. He has held visiting positions at UCLA. Before coming to McGill in 2012, he was tenured at Simon Fraser university, and a postdoc at University of Texas, Austin. He was a student at the University of Toronto (undergraduate) and University of Chicago (graduate).

His research focusses on mathematical approaches to machine learning: optimization (including stochastic gradient descent), regularization approaches (including averaged models and gradient regularization), and robust models, including adversarially trained models. He teaches a theory course on machine learning, including generalization theory, and a scientific computing course with a focus on high dimensional methods.

Publications

2021-03

Scaleable input gradient regularization for adversarial robustness
Chris Finlay and Adam M. Oberman
Machine Learning with Applications
(2021-03-01)
dblp.uni-trier.dePDF

2020-10

Adversarial Boot Camp: label free certified robustness in one epoch.
Ryan Campbell, Chris Finlay and Adam M Oberman
arXiv preprint arXiv:2010.02508
(2020-10-05)
arxiv.orgPDF

2020-09

Uncertainty for deep image classifiers on out of distribution data.
Tiago Salvador, Alexander Iannantuono and Adam M Oberman
(venue unknown)
(2020-09-28)
openreview.netPDF

2020-07

How to train your Neural ODE
Chris Finlay, Joern-Henrik Jacobsen, Levon Nurbekyan and Adam M Oberman
ICML 2020
(2020-07-12)
icml.ccPDF

2020-06

Learning normalizing flows from Entropy-Kantorovich potentials
Chris Finlay, Augusto Gerolin, Adam M Oberman and Aram-Alexandre Pooladian
arXiv preprint arXiv:2006.06033
(2020-06-10)
ui.adsabs.harvard.eduPDF
Deterministic Gaussian Averaged Neural Networks.
Ryan Campbell, Chris Finlay and Adam M. Oberman
arXiv: Learning
(2020-06-10)
dblp.uni-trier.dePDF
A principled approach for generating adversarial images under non-smooth dissimilarity metrics
Aram-Alexandre Pooladian, Chris Finlay, Tim Hoheisel and Adam M. Oberman
AISTATS 2020
(2020-06-03)
proceedings.mlr.pressPDF

Publications collected and formatted using Paperoni

array(1) { ["wp-wpml_current_language"]=> string(2) "fr" }