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 Behnam Neyshabur

Behnam Neyshabur

Google Research

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

* 20 pages, 2 figures. Accepted for NeurIPS 2020 Competitions Track. Lead organizer: Yiding Jiang 

  Access Paper or Ask Questions

When Do Curricula Work?


Dec 05, 2020
Xiaoxia Wu, Ethan Dyer, Behnam Neyshabur

* ICLR 2021 

  Access Paper or Ask Questions

Understanding the Failure Modes of Out-of-Distribution Generalization


Oct 29, 2020
Vaishnavh Nagarajan, Anders Andreassen, Behnam Neyshabur


  Access Paper or Ask Questions

Are wider nets better given the same number of parameters?


Oct 27, 2020
Anna Golubeva, Behnam Neyshabur, Guy Gur-Ari

* 9 pages 

  Access Paper or Ask Questions

The Deep Bootstrap: Good Online Learners are Good Offline Generalizers


Oct 16, 2020
Preetum Nakkiran, Behnam Neyshabur, Hanie Sedghi


  Access Paper or Ask Questions

Sharpness-Aware Minimization for Efficiently Improving Generalization


Oct 03, 2020
Pierre Foret, Ariel Kleiner, Hossein Mobahi, Behnam Neyshabur


  Access Paper or Ask Questions

Extreme Memorization via Scale of Initialization


Aug 31, 2020
Harsh Mehta, Ashok Cutkosky, Behnam Neyshabur


  Access Paper or Ask Questions

What is being transferred in transfer learning?


Aug 26, 2020
Behnam Neyshabur, Hanie Sedghi, Chiyuan Zhang

* Equal contribution, authors ordered randomly 

  Access Paper or Ask Questions

Towards Learning Convolutions from Scratch


Jul 27, 2020
Behnam Neyshabur

* 18 pages, 9 figures, 4 tables 

  Access Paper or Ask Questions

Observational Overfitting in Reinforcement Learning


Dec 28, 2019
Xingyou Song, Yiding Jiang, Stephen Tu, Yilun Du, Behnam Neyshabur

* Published as a conference paper in ICLR 2020 

  Access Paper or Ask Questions

The intriguing role of module criticality in the generalization of deep networks


Dec 04, 2019
Niladri S. Chatterji, Behnam Neyshabur, Hanie Sedghi


  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


  Access Paper or Ask Questions

Stronger generalization bounds for deep nets via a compression approach


Nov 05, 2018
Sanjeev Arora, Rong Ge, Behnam Neyshabur, Yi Zhang


  Access Paper or Ask Questions

Stabilizing GAN Training with Multiple Random Projections


Jun 23, 2018
Behnam Neyshabur, Srinadh Bhojanapalli, Ayan Chakrabarti


  Access Paper or Ask Questions

Towards Understanding the Role of Over-Parametrization in Generalization of Neural Networks


May 30, 2018
Behnam Neyshabur, Zhiyuan Li, Srinadh Bhojanapalli, Yann LeCun, Nathan Srebro

* 19 pages, 8 figures 

  Access Paper or Ask Questions

A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks


Feb 23, 2018
Behnam Neyshabur, Srinadh Bhojanapalli, Nathan Srebro

* Accepted to ICLR 2018 

  Access Paper or Ask Questions

Implicit Regularization in Deep Learning


Sep 08, 2017
Behnam Neyshabur

* PhD Thesis 

  Access Paper or Ask Questions

Exploring Generalization in Deep Learning


Jul 06, 2017
Behnam Neyshabur, Srinadh Bhojanapalli, David McAllester, Nathan Srebro

* 19 pages, 8 figures 

  Access Paper or Ask Questions

Corralling a Band of Bandit Algorithms


Jun 06, 2017
Alekh Agarwal, Haipeng Luo, Behnam Neyshabur, Robert E. Schapire

* Accepted to COLT 2017 

  Access Paper or Ask Questions

Implicit Regularization in Matrix Factorization


May 25, 2017
Suriya Gunasekar, Blake Woodworth, Srinadh Bhojanapalli, Behnam Neyshabur, Nathan Srebro


  Access Paper or Ask Questions

Geometry of Optimization and Implicit Regularization in Deep Learning


May 08, 2017
Behnam Neyshabur, Ryota Tomioka, Ruslan Salakhutdinov, Nathan Srebro

* This survey chapter was done as a part of Intel Collaborative Research institute for Computational Intelligence (ICRI-CI) "Why & When Deep Learning works -- looking inside Deep Learning" compendium with the generous support of ICRI-CI. arXiv admin note: substantial text overlap with arXiv:1506.02617 

  Access Paper or Ask Questions

Global Optimality of Local Search for Low Rank Matrix Recovery


May 27, 2016
Srinadh Bhojanapalli, Behnam Neyshabur, Nathan Srebro

* 21 pages, 3 figures 

  Access Paper or Ask Questions

Path-Normalized Optimization of Recurrent Neural Networks with ReLU Activations


May 23, 2016
Behnam Neyshabur, Yuhuai Wu, Ruslan Salakhutdinov, Nathan Srebro

* 15 pages 

  Access Paper or Ask Questions

Data-Dependent Path Normalization in Neural Networks


Jan 19, 2016
Behnam Neyshabur, Ryota Tomioka, Ruslan Salakhutdinov, Nathan Srebro

* 17 pages, 3 figures 

  Access Paper or Ask Questions

On Symmetric and Asymmetric LSHs for Inner Product Search


Jun 08, 2015
Behnam Neyshabur, Nathan Srebro

* 11 pages, 3 figures, In Proceedings of The 32nd International Conference on Machine Learning (ICML) 

  Access Paper or Ask Questions

Path-SGD: Path-Normalized Optimization in Deep Neural Networks


Jun 08, 2015
Behnam Neyshabur, Ruslan Salakhutdinov, Nathan Srebro

* 12 pages, 5 figures 

  Access Paper or Ask Questions

In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning


Apr 16, 2015
Behnam Neyshabur, Ryota Tomioka, Nathan Srebro

* 9 pages, 2 figures 

  Access Paper or Ask Questions

Norm-Based Capacity Control in Neural Networks


Apr 14, 2015
Behnam Neyshabur, Ryota Tomioka, Nathan Srebro

* 29 pages 

  Access Paper or Ask Questions