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
Book ChapterDOI
Spectral Dimensionality Reduction
Yoshua Bengio,Olivier Delalleau,Nicolas Le Roux,Jean-François Paiement,Pascal Vincent,Marie Claude Ouimet +5 more
TL;DR: A number of non-linear dimensionality reduction methods, such as Locally Linear Embedding, Isomap, Laplacian Eigenmaps and kernel PCA, which are based on performing an eigen-decomposition are put under a common framework.
Posted Content
DEFactor: Differentiable Edge Factorization-based Probabilistic Graph Generation
TL;DR: This work proposes a model for conditional graph generation that is computationally efficient and enables direct optimisation of the graph and demonstrates favourable performance of the model on prototype-based molecular graph conditional generation tasks.
Posted Content
Diet Networks: Thin Parameters for Fat Genomics
Adriana Romero,Pierre Luc Carrier,Akram Erraqabi,Tristan Sylvain,Alex Auvolat,Etienne Dejoie,Marc-André Legault,Marie-Pierre Dubé,Julie Hussin,Yoshua Bengio +9 more
TL;DR: In this paper, the authors proposed a novel neural network parametrization which considerably reduces the number of free parameters in the first layer of a classifier neural network by learning or providing a distributed representation for each input feature (e.g. for each position in the genome where variations are observed).
Posted Content
On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length
Stanisław Jastrzębski,Stanisław Jastrzębski,Zachary Kenton,Nicolas Ballas,Asja Fischer,Yoshua Bengio,Yoshua Bengio,Amos Storkey +7 more
TL;DR: This article studied the SGD dynamics in relation to the sharpest directions in the initial phase and found that SGD step is large compared to the curvature and commonly fails to minimize the loss along these directions.
Proceedings Article
On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length
Stanisław Jastrzębski,Stanisław Jastrzębski,Zachary Kenton,Nicolas Ballas,Asja Fischer,Yoshua Bengio,Yoshua Bengio,Amos Storkey +7 more
TL;DR: The authors studied the SGD dynamics in relation to the sharpest directions in the initial phase and found that SGD step is large compared to the curvature and commonly fails to minimize the loss along these directions.