Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Samy Bengio

Samy Bengio

Google Research

Training cascaded networks for speeded decisions using a temporal-difference loss


Feb 19, 2021
Michael L. Iuzzolino, Michael C. Mozer, Samy Bengio

Figure 1 for Training cascaded networks for speeded decisions using a temporal-difference loss
Figure 2 for Training cascaded networks for speeded decisions using a temporal-difference loss
Figure 3 for Training cascaded networks for speeded decisions using a temporal-difference loss
Figure 4 for Training cascaded networks for speeded decisions using a temporal-difference loss

  Access Paper or Ask Questions

NeurIPS 2020 Competition: Predicting Generalization in Deep Learning


Dec 14, 2020
Yiding Jiang, Pierre Foret, Scott Yak, Daniel M. Roy, Hossein Mobahi, Gintare Karolina Dziugaite, Samy Bengio, Suriya Gunasekar, Isabelle Guyon, Behnam Neyshabur

Figure 1 for NeurIPS 2020 Competition: Predicting Generalization in Deep Learning
* 20 pages, 2 figures. Accepted for NeurIPS 2020 Competitions Track. Lead organizer: Yiding Jiang 
  Access Paper or Ask Questions

Data Augmentation via Structured Adversarial Perturbations


Nov 05, 2020
Calvin Luo, Hossein Mobahi, Samy Bengio

Figure 1 for Data Augmentation via Structured Adversarial Perturbations
Figure 2 for Data Augmentation via Structured Adversarial Perturbations
Figure 3 for Data Augmentation via Structured Adversarial Perturbations
Figure 4 for Data Augmentation via Structured Adversarial Perturbations

  Access Paper or Ask Questions

Characterising Bias in Compressed Models


Oct 06, 2020
Sara Hooker, Nyalleng Moorosi, Gregory Clark, Samy Bengio, Emily Denton

Figure 1 for Characterising Bias in Compressed Models
Figure 2 for Characterising Bias in Compressed Models
Figure 3 for Characterising Bias in Compressed Models
Figure 4 for Characterising Bias in Compressed Models

  Access Paper or Ask Questions

Auto Completion of User Interface Layout Design Using Transformer-Based Tree Decoders


Jan 14, 2020
Yang Li, Julien Amelot, Xin Zhou, Samy Bengio, Si Si

Figure 1 for Auto Completion of User Interface Layout Design Using Transformer-Based Tree Decoders
Figure 2 for Auto Completion of User Interface Layout Design Using Transformer-Based Tree Decoders
Figure 3 for Auto Completion of User Interface Layout Design Using Transformer-Based Tree Decoders
Figure 4 for Auto Completion of User Interface Layout Design Using Transformer-Based Tree Decoders

  Access Paper or Ask Questions

Fantastic Generalization Measures and Where to Find Them


Dec 04, 2019
Yiding Jiang, Behnam Neyshabur, Hossein Mobahi, Dilip Krishnan, Samy Bengio

Figure 1 for Fantastic Generalization Measures and Where to Find Them
Figure 2 for Fantastic Generalization Measures and Where to Find Them
Figure 3 for Fantastic Generalization Measures and Where to Find Them
Figure 4 for Fantastic Generalization Measures and Where to Find Them

  Access Paper or Ask Questions

Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML


Sep 19, 2019
Aniruddh Raghu, Maithra Raghu, Samy Bengio, Oriol Vinyals

Figure 1 for Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML
Figure 2 for Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML
Figure 3 for Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML
Figure 4 for Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML

  Access Paper or Ask Questions

Efficient Exploration with Self-Imitation Learning via Trajectory-Conditioned Policy


Jul 24, 2019
Yijie Guo, Jongwook Choi, Marcin Moczulski, Samy Bengio, Mohammad Norouzi, Honglak Lee

Figure 1 for Efficient Exploration with Self-Imitation Learning via Trajectory-Conditioned Policy
Figure 2 for Efficient Exploration with Self-Imitation Learning via Trajectory-Conditioned Policy
Figure 3 for Efficient Exploration with Self-Imitation Learning via Trajectory-Conditioned Policy
Figure 4 for Efficient Exploration with Self-Imitation Learning via Trajectory-Conditioned Policy

  Access Paper or Ask Questions

Parallel Scheduled Sampling


Jun 11, 2019
Daniel Duckworth, Arvind Neelakantan, Ben Goodrich, Lukasz Kaiser, Samy Bengio

Figure 1 for Parallel Scheduled Sampling
Figure 2 for Parallel Scheduled Sampling
Figure 3 for Parallel Scheduled Sampling
Figure 4 for Parallel Scheduled Sampling
* Initial submission 
  Access Paper or Ask Questions

A Closed-Form Learned Pooling for Deep Classification Networks


Jun 10, 2019
Vighnesh Birodkar, Hossein Mobahi, Dilip Krishnan, Samy Bengio

Figure 1 for A Closed-Form Learned Pooling for Deep Classification Networks
Figure 2 for A Closed-Form Learned Pooling for Deep Classification Networks
Figure 3 for A Closed-Form Learned Pooling for Deep Classification Networks
Figure 4 for A Closed-Form Learned Pooling for Deep Classification Networks

  Access Paper or Ask Questions

Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks


May 20, 2019
Wei-Lin Chiang, Xuanqing Liu, Si Si, Yang Li, Samy Bengio, Cho-Jui Hsieh

Figure 1 for Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
Figure 2 for Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
Figure 3 for Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
Figure 4 for Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks
* In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD'19) 
  Access Paper or Ask Questions

Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure


Mar 21, 2019
Been Kim, Emily Reif, Martin Wattenberg, Samy Bengio

Figure 1 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure
Figure 2 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure
Figure 3 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure
Figure 4 for Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

  Access Paper or Ask Questions

Identity Crisis: Memorization and Generalization under Extreme Overparameterization


Feb 15, 2019
Chiyuan Zhang, Samy Bengio, Moritz Hardt, Yoram Singer

Figure 1 for Identity Crisis: Memorization and Generalization under Extreme Overparameterization
Figure 2 for Identity Crisis: Memorization and Generalization under Extreme Overparameterization
Figure 3 for Identity Crisis: Memorization and Generalization under Extreme Overparameterization
Figure 4 for Identity Crisis: Memorization and Generalization under Extreme Overparameterization

  Access Paper or Ask Questions

Transfusion: Understanding Transfer Learning with Applications to Medical Imaging


Feb 14, 2019
Maithra Raghu, Chiyuan Zhang, Jon Kleinberg, Samy Bengio

Figure 1 for Transfusion: Understanding Transfer Learning with Applications to Medical Imaging
Figure 2 for Transfusion: Understanding Transfer Learning with Applications to Medical Imaging
Figure 3 for Transfusion: Understanding Transfer Learning with Applications to Medical Imaging
Figure 4 for Transfusion: Understanding Transfer Learning with Applications to Medical Imaging

  Access Paper or Ask Questions

Are All Layers Created Equal?


Feb 12, 2019
Chiyuan Zhang, Samy Bengio, Yoram Singer

Figure 1 for Are All Layers Created Equal?
Figure 2 for Are All Layers Created Equal?
Figure 3 for Are All Layers Created Equal?
Figure 4 for Are All Layers Created Equal?

  Access Paper or Ask Questions

Semantic Redundancies in Image-Classification Datasets: The 10% You Don't Need


Jan 29, 2019
Vighnesh Birodkar, Hossein Mobahi, Samy Bengio

Figure 1 for Semantic Redundancies in Image-Classification Datasets: The 10% You Don't Need
Figure 2 for Semantic Redundancies in Image-Classification Datasets: The 10% You Don't Need
Figure 3 for Semantic Redundancies in Image-Classification Datasets: The 10% You Don't Need
Figure 4 for Semantic Redundancies in Image-Classification Datasets: The 10% You Don't Need

  Access Paper or Ask Questions

Unsupervised speech representation learning using WaveNet autoencoders


Jan 25, 2019
Jan Chorowski, Ron J. Weiss, Samy Bengio, Aäron van den Oord

Figure 1 for Unsupervised speech representation learning using WaveNet autoencoders
Figure 2 for Unsupervised speech representation learning using WaveNet autoencoders
Figure 3 for Unsupervised speech representation learning using WaveNet autoencoders
Figure 4 for Unsupervised speech representation learning using WaveNet autoencoders

  Access Paper or Ask Questions

GaterNet: Dynamic Filter Selection in Convolutional Neural Network via a Dedicated Global Gating Network


Nov 27, 2018
Zhourong Chen, Yang Li, Samy Bengio, Si Si

Figure 1 for GaterNet: Dynamic Filter Selection in Convolutional Neural Network via a Dedicated Global Gating Network
Figure 2 for GaterNet: Dynamic Filter Selection in Convolutional Neural Network via a Dedicated Global Gating Network
Figure 3 for GaterNet: Dynamic Filter Selection in Convolutional Neural Network via a Dedicated Global Gating Network
Figure 4 for GaterNet: Dynamic Filter Selection in Convolutional Neural Network via a Dedicated Global Gating Network
* Google Research 
  Access Paper or Ask Questions

Content preserving text generation with attribute controls


Nov 03, 2018
Lajanugen Logeswaran, Honglak Lee, Samy Bengio

Figure 1 for Content preserving text generation with attribute controls
Figure 2 for Content preserving text generation with attribute controls
Figure 3 for Content preserving text generation with attribute controls
Figure 4 for Content preserving text generation with attribute controls
* NIPS 2018 
  Access Paper or Ask Questions

Area Attention


Oct 30, 2018
Yang Li, Lukasz Kaiser, Samy Bengio, Si Si

Figure 1 for Area Attention
Figure 2 for Area Attention
Figure 3 for Area Attention
Figure 4 for Area Attention

  Access Paper or Ask Questions

Insights on representational similarity in neural networks with canonical correlation


Oct 23, 2018
Ari S. Morcos, Maithra Raghu, Samy Bengio

Figure 1 for Insights on representational similarity in neural networks with canonical correlation
Figure 2 for Insights on representational similarity in neural networks with canonical correlation
Figure 3 for Insights on representational similarity in neural networks with canonical correlation
Figure 4 for Insights on representational similarity in neural networks with canonical correlation
* NIPS 2018 
  Access Paper or Ask Questions

Predicting the Generalization Gap in Deep Networks with Margin Distributions


Sep 28, 2018
Yiding Jiang, Dilip Krishnan, Hossein Mobahi, Samy Bengio

Figure 1 for Predicting the Generalization Gap in Deep Networks with Margin Distributions
Figure 2 for Predicting the Generalization Gap in Deep Networks with Margin Distributions
Figure 3 for Predicting the Generalization Gap in Deep Networks with Margin Distributions
Figure 4 for Predicting the Generalization Gap in Deep Networks with Margin Distributions

  Access Paper or Ask Questions

Time-Dependent Representation for Neural Event Sequence Prediction


Jul 20, 2018
Yang Li, Nan Du, Samy Bengio

Figure 1 for Time-Dependent Representation for Neural Event Sequence Prediction
Figure 2 for Time-Dependent Representation for Neural Event Sequence Prediction
Figure 3 for Time-Dependent Representation for Neural Event Sequence Prediction
Figure 4 for Time-Dependent Representation for Neural Event Sequence Prediction
* 9 pages and 2 pages of references 
  Access Paper or Ask Questions

Fast Decoding in Sequence Models using Discrete Latent Variables


Jun 07, 2018
Łukasz Kaiser, Aurko Roy, Ashish Vaswani, Niki Parmar, Samy Bengio, Jakob Uszkoreit, Noam Shazeer

Figure 1 for Fast Decoding in Sequence Models using Discrete Latent Variables
Figure 2 for Fast Decoding in Sequence Models using Discrete Latent Variables
Figure 3 for Fast Decoding in Sequence Models using Discrete Latent Variables
Figure 4 for Fast Decoding in Sequence Models using Discrete Latent Variables
* ICML 2018 
  Access Paper or Ask Questions

A Study on Overfitting in Deep Reinforcement Learning


Apr 20, 2018
Chiyuan Zhang, Oriol Vinyals, Remi Munos, Samy Bengio

Figure 1 for A Study on Overfitting in Deep Reinforcement Learning
Figure 2 for A Study on Overfitting in Deep Reinforcement Learning
Figure 3 for A Study on Overfitting in Deep Reinforcement Learning
Figure 4 for A Study on Overfitting in Deep Reinforcement Learning

  Access Paper or Ask Questions

Adversarial Attacks and Defences Competition


Mar 31, 2018
Alexey Kurakin, Ian Goodfellow, Samy Bengio, Yinpeng Dong, Fangzhou Liao, Ming Liang, Tianyu Pang, Jun Zhu, Xiaolin Hu, Cihang Xie, Jianyu Wang, Zhishuai Zhang, Zhou Ren, Alan Yuille, Sangxia Huang, Yao Zhao, Yuzhe Zhao, Zhonglin Han, Junjiajia Long, Yerkebulan Berdibekov, Takuya Akiba, Seiya Tokui, Motoki Abe

Figure 1 for Adversarial Attacks and Defences Competition
Figure 2 for Adversarial Attacks and Defences Competition
Figure 3 for Adversarial Attacks and Defences Competition
Figure 4 for Adversarial Attacks and Defences Competition
* 36 pages, 10 figures 
  Access Paper or Ask Questions

Tensor2Tensor for Neural Machine Translation


Mar 16, 2018
Ashish Vaswani, Samy Bengio, Eugene Brevdo, Francois Chollet, Aidan N. Gomez, Stephan Gouws, Llion Jones, Łukasz Kaiser, Nal Kalchbrenner, Niki Parmar, Ryan Sepassi, Noam Shazeer, Jakob Uszkoreit

Figure 1 for Tensor2Tensor for Neural Machine Translation
* arXiv admin note: text overlap with arXiv:1706.03762 
  Access Paper or Ask Questions

Large Margin Deep Networks for Classification


Mar 15, 2018
Gamaleldin F. Elsayed, Dilip Krishnan, Hossein Mobahi, Kevin Regan, Samy Bengio

Figure 1 for Large Margin Deep Networks for Classification
Figure 2 for Large Margin Deep Networks for Classification
Figure 3 for Large Margin Deep Networks for Classification
Figure 4 for Large Margin Deep Networks for Classification

  Access Paper or Ask Questions

On Using Backpropagation for Speech Texture Generation and Voice Conversion


Mar 08, 2018
Jan Chorowski, Ron J. Weiss, Rif A. Saurous, Samy Bengio

Figure 1 for On Using Backpropagation for Speech Texture Generation and Voice Conversion
Figure 2 for On Using Backpropagation for Speech Texture Generation and Voice Conversion
Figure 3 for On Using Backpropagation for Speech Texture Generation and Voice Conversion
Figure 4 for On Using Backpropagation for Speech Texture Generation and Voice Conversion
* Accepted to ICASSP 2018 
  Access Paper or Ask Questions