Y
Yoshua Bengio
Researcher at Université de Montréal
Publications - 1146
Citations - 534376
Yoshua Bengio is an academic researcher from Université de Montréal. The author has contributed to research in topics: Artificial neural network & Deep learning. The author has an hindex of 202, co-authored 1033 publications receiving 420313 citations. Previous affiliations of Yoshua Bengio include McGill University & Centre de Recherches Mathématiques.
Papers
More filters
Journal ArticleDOI
MixupE: Understanding and Improving Mixup from Directional Derivative Perspective
Vikas Verma,Sarthak Mittal,Wai Hoh Tang,Hieu Quang Pham,Juho Kannala,Yoshua Bengio,Arno Solin,Kenji Kawaguchi +7 more
TL;DR: In this paper , the authors proposed an improved version of Mixup, theoretically justified to deliver better generalization performance than the vanilla mixup, which implicitly regularizes infinitely many directional derivatives of all orders.
Journal ArticleDOI
Lookback for Learning to Branch
Prateek Gupta,Elias B. Khalil,Didier Chet'elat,Maxime Gasse,Yoshua Bengio,Andrea Lodi,M. Pawan Kumar +6 more
TL;DR: To imitate the target behavior more closely by incorporating the lookback phenomenon in GNNs, two methods are proposed: (a) target smoothing for the standard cross-entropy loss function, and (b) adding a Parent-as-Target (PAT) Lookback regularizer term.
Posted Content
Fast and Slow Learning of Recurrent Independent Mechanisms
TL;DR: In this paper, an attention mechanism dynamically selects which modules can be adapted to the current task, and the parameters of the selected modules are allowed to change quickly as the learner is confronted with variations in what it experiences.
Journal ArticleDOI
Agnostic Physics-Driven Deep Learning
TL;DR: This work establishes that a physical system can perform statistical learning without gradient computations, via an Agnostic Equilibrium Propagation procedure that combines energy minimization, homeostatic control, and nudging towards the correct response.
Proceedings Article
Meta Attention Networks: Meta-Learning Attention to Modulate Information Between Recurrent Independent Mechanisms
TL;DR: In this article, the authors propose a particular training framework in which the pieces of knowledge an agent needs, as well as its reward function are stationary and can be re-used across tasks, and find that meta-learning the modular aspects of the proposed system greatly help in achieving faster learning, in experiments with a reinforcement learning setup involving navigation in a partially observed grid world with image-level input.