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

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

SGD learning on neural networks: leap complexity and saddle-to-saddle dynamics


Feb 21, 2023
Emmanuel Abbe, Enric Boix-Adsera, Theodor Misiakiewicz

Add code


   Access Paper or Ask Questions

Generalization on the Unseen, Logic Reasoning and Degree Curriculum


Jan 30, 2023
Emmanuel Abbe, Samy Bengio, Aryo Lotfi, Kevin Rizk

Add code

* 37 pages, 10 figures 

   Access Paper or Ask Questions

On the non-universality of deep learning: quantifying the cost of symmetry


Aug 05, 2022
Emmanuel Abbe, Enric Boix-Adsera

Add code


   Access Paper or Ask Questions

Learning to Reason with Neural Networks: Generalization, Unseen Data and Boolean Measures


May 26, 2022
Emmanuel Abbe, Samy Bengio, Elisabetta Cornacchia, Jon Kleinberg, Aryo Lotfi, Maithra Raghu, Chiyuan Zhang

Add code

* 28 pages, 8 figures 

   Access Paper or Ask Questions

An initial alignment between neural network and target is needed for gradient descent to learn


Feb 25, 2022
Emmanuel Abbe, Elisabetta Cornacchia, Jan Hązła, Christopher Marquis

Add code


   Access Paper or Ask Questions

The merged-staircase property: a necessary and nearly sufficient condition for SGD learning of sparse functions on two-layer neural networks


Feb 17, 2022
Emmanuel Abbe, Enric Boix-Adsera, Theodor Misiakiewicz

Add code


   Access Paper or Ask Questions

Binary perceptron: efficient algorithms can find solutions in a rare well-connected cluster


Nov 04, 2021
Emmanuel Abbe, Shuangping Li, Allan Sly

Add code


   Access Paper or Ask Questions

The staircase property: How hierarchical structure can guide deep learning


Aug 24, 2021
Emmanuel Abbe, Enric Boix-Adsera, Matthew Brennan, Guy Bresler, Dheeraj Nagaraj

Add code


   Access Paper or Ask Questions

On the Power of Differentiable Learning versus PAC and SQ Learning


Aug 09, 2021
Emmanuel Abbe, Pritish Kamath, Eran Malach, Colin Sandon, Nathan Srebro

Add code


   Access Paper or Ask Questions

Quantifying the Benefit of Using Differentiable Learning over Tangent Kernels


Mar 01, 2021
Eran Malach, Pritish Kamath, Emmanuel Abbe, Nathan Srebro

Add code


   Access Paper or Ask Questions

1
2
3
>>