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 Chulhee Yun

Minibatch vs Local SGD with Shuffling: Tight Convergence Bounds and Beyond


Oct 20, 2021
Chulhee Yun, Shashank Rajput, Suvrit Sra

* 72 pages 

  Access Paper or Ask Questions

Can Single-Shuffle SGD be Better than Reshuffling SGD and GD?


Mar 12, 2021
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 26 pages, 2 figures 

  Access Paper or Ask Questions

Provable Memorization via Deep Neural Networks using Sub-linear Parameters


Oct 26, 2020
Sejun Park, Jaeho Lee, Chulhee Yun, Jinwoo Shin


  Access Paper or Ask Questions

A Unifying View on Implicit Bias in Training Linear Neural Networks


Oct 06, 2020
Chulhee Yun, Shankar Krishnan, Hossein Mobahi

* 40 pages, 6 figures 

  Access Paper or Ask Questions

SGD with shuffling: optimal rates without component convexity and large epoch requirements


Jun 22, 2020
Kwangjun Ahn, Chulhee Yun, Suvrit Sra

* 53 pages; supersedes the preprint arXiv:2004.08657; v2 corrects an erroneous claim about SingleShuffle and newly adds Theorem 24 and Appendix F for SingleShuffle 

  Access Paper or Ask Questions

Minimum Width for Universal Approximation


Jun 16, 2020
Sejun Park, Chulhee Yun, Jaeho Lee, Jinwoo Shin


  Access Paper or Ask Questions

$O(n)$ Connections are Expressive Enough: Universal Approximability of Sparse Transformers


Jun 08, 2020
Chulhee Yun, Yin-Wen Chang, Srinadh Bhojanapalli, Ankit Singh Rawat, Sashank J. Reddi, Sanjiv Kumar

* 32 pages 

  Access Paper or Ask Questions

Low-Rank Bottleneck in Multi-head Attention Models


Feb 17, 2020
Srinadh Bhojanapalli, Chulhee Yun, Ankit Singh Rawat, Sashank J. Reddi, Sanjiv Kumar

* 17 pages, 4 figures 

  Access Paper or Ask Questions

Are Transformers universal approximators of sequence-to-sequence functions?


Dec 20, 2019
Chulhee Yun, Srinadh Bhojanapalli, Ankit Singh Rawat, Sashank J. Reddi, Sanjiv Kumar

* Accepted to ICLR 2020 

  Access Paper or Ask Questions

Are deep ResNets provably better than linear predictors?


Jul 09, 2019
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 17 pages 

  Access Paper or Ask Questions

Finite sample expressive power of small-width ReLU networks


Oct 17, 2018
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 17 pages, 2 figures 

  Access Paper or Ask Questions

Efficiently testing local optimality and escaping saddles for ReLU networks


Sep 28, 2018
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 25 pages 

  Access Paper or Ask Questions

Small nonlinearities in activation functions create bad local minima in neural networks


Sep 28, 2018
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 36 pages 

  Access Paper or Ask Questions

Global optimality conditions for deep neural networks


Mar 24, 2018
Chulhee Yun, Suvrit Sra, Ali Jadbabaie

* 14 pages. A camera-ready version that will appear at ICLR 2018 

  Access Paper or Ask Questions