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

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Demystifying Disagreement-on-the-Line in High Dimensions


Jan 31, 2023
Donghwan Lee, Behrad Moniri, Xinmeng Huang, Edgar Dobriban, Hamed Hassani

Add code


   Access Paper or Ask Questions

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

Optimal Complexity in Non-Convex Decentralized Learning over Time-Varying Networks


Nov 01, 2022
Xinmeng Huang, Kun Yuan

Add code

* Accepted by 14th Annual Workshop on Optimization for Machine Learning. arXiv admin note: text overlap with arXiv:2210.07863 

   Access Paper or Ask Questions

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

Revisiting Optimal Convergence Rate for Smooth and Non-convex Stochastic Decentralized Optimization


Oct 14, 2022
Kun Yuan, Xinmeng Huang, Yiming Chen, Xiaohan Zhang, Yingya Zhang, Pan Pan

Add code


   Access Paper or Ask Questions

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

Lower Bounds and Nearly Optimal Algorithms in Distributed Learning with Communication Compression


Jun 08, 2022
Xinmeng Huang, Yiming Chen, Wotao Yin, Kun Yuan

Add code


   Access Paper or Ask Questions

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

Collaborative Learning of Distributions under Heterogeneity and Communication Constraints


Jun 07, 2022
Xinmeng Huang, Donghwan Lee, Edgar Dobriban, Hamed Hassani

Add code


   Access Paper or Ask Questions

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

T-Cal: An optimal test for the calibration of predictive models


Mar 25, 2022
Donghwan Lee, Xinmeng Huang, Hamed Hassani, Edgar Dobriban

Add code

* The implementation of T-Cal is available at https://github.com/dh7401/T-Cal 

   Access Paper or Ask Questions

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

On the Comparison between Cyclic Sampling and Random Reshuffling


Apr 25, 2021
Xinmeng Huang, Kun Yuan, Xianghui Mao, Wotao Yin

Add code


   Access Paper or Ask Questions

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

DecentLaM: Decentralized Momentum SGD for Large-batch Deep Training


Apr 24, 2021
Kun Yuan, Yiming Chen, Xinmeng Huang, Yingya Zhang, Pan Pan, Yinghui Xu, Wotao Yin

Add code


   Access Paper or Ask Questions

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