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

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Neural Networks Efficiently Learn Low-Dimensional Representations with SGD


Sep 29, 2022
Alireza Mousavi-Hosseini, Sejun Park, Manuela Girotti, Ioannis Mitliagkas, Murat A. Erdogdu


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Generalization Bounds for Stochastic Gradient Descent via Localized $\varepsilon$-Covers


Sep 19, 2022
Sejun Park, Umut ƞimƟekli, Murat A. Erdogdu


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

$p$-DkNN: Out-of-Distribution Detection Through Statistical Testing of Deep Representations


Jul 25, 2022
Adam Dziedzic, Stephan Rabanser, Mohammad Yaghini, Armin Ale, Murat A. Erdogdu, Nicolas Papernot


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation


May 03, 2022
Jimmy Ba, Murat A. Erdogdu, Taiji Suzuki, Zhichao Wang, Denny Wu, Greg Yang

* 71 pages 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Mirror Descent Strikes Again: Optimal Stochastic Convex Optimization under Infinite Noise Variance


Feb 23, 2022
Nuri Mert Vural, Lu Yu, Krishnakumar Balasubramanian, Stanislav Volgushev, Murat A. Erdogdu

* 31 pages, 1 figure 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Towards a Theory of Non-Log-Concave Sampling: First-Order Stationarity Guarantees for Langevin Monte Carlo


Feb 10, 2022
Krishnakumar Balasubramanian, Sinho Chewi, Murat A. Erdogdu, Adil Salim, Matthew Zhang


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Heavy-tailed Sampling via Transformed Unadjusted Langevin Algorithm


Jan 20, 2022
Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Analysis of Langevin Monte Carlo from Poincaré to Log-Sobolev


Dec 23, 2021
Sinho Chewi, Murat A. Erdogdu, Mufan Bill Li, Ruoqi Shen, Matthew Zhang

* 35 pages 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

On Empirical Risk Minimization with Dependent and Heavy-Tailed Data


Sep 10, 2021
Abhishek Roy, Krishnakumar Balasubramanian, Murat A. Erdogdu

* fixed minor typos 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email
1
2
3
>>