Chulhee Yun

发表

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

Sashank J. Reddi, Srinadh Bhojanapalli, Sanjiv Kumar, 2020, NeurIPS.

Global optimality conditions for deep neural networks pdf

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2017, ICLR.

Minimum Width for Universal Approximation pdf

Jinwoo Shin, Jaeho Lee, Chulhee Yun, 2020, ICLR.

Finite sample expressive power of small-width ReLU networks

A. Jadbabaie, S. Sra, Chulhee Yun, 2018, ArXiv.

LOBAL OPTIMALITY CONDITIONS FOR DEEP NEURAL NETWORKS

A. Jadbabaie, S. Sra, Chulhee Yun, 2018 .

Concise Multi-head Attention Models

Ankit Singh Rawat, Sashank J. Reddi, Sanjiv Kumar, 2019 .

Provable Memorization via Deep Neural Networks using Sub-linear Parameters

Jinwoo Shin, Jaeho Lee, Chulhee Yun, 2020, COLT.

SGDA with shuffling: faster convergence for nonconvex-PŁ minimax optimization

Chulhee Yun, Hanseul Cho, 2022, International Conference on Learning Representations.

Small ReLU networks are powerful memorizers: a tight analysis of memorization capacity

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2018, NeurIPS.

Are deep ResNets provably better than linear predictors?

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2019, NeurIPS.

Efficiently testing local optimality and escaping saddles for ReLU networks pdf

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2018, ICLR.

Minimax Bounds on Stochastic Batched Convex Optimization

Feng Ruan, John C. Duchi, Chulhee Yun, 2018, COLT.

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

Sashank J. Reddi, Srinadh Bhojanapalli, Sanjiv Kumar, 2019, ICLR.

Low-Rank Bottleneck in Multi-head Attention Models pdf

Ankit Singh Rawat, Sashank J. Reddi, Sanjiv Kumar, 2020, ICML.

A Unifying View on Implicit Bias in Training Linear Neural Networks

H. Mobahi, Chulhee Yun, Shankar Krishnan, 2020, ICLR.

Tighter Lower Bounds for Shuffling SGD: Random Permutations and Beyond

Chulhee Yun, Jaewook Lee, Jaeyoung Cha, 2023, ArXiv.

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

Suvrit Sra, Kwangjun Ahn, Chulhee Yun, 2020, NeurIPS.

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

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2018, ICLR.

Practical Sharpness-Aware Minimization Cannot Converge All the Way to Optima

Chulhee Yun, Dongkuk Si, 2023, ArXiv.

SGDA with shuffling: faster convergence for nonconvex-PŁ minimax optimization

Chulhee Yun, Hanseul Cho, 2022, ICLR.

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

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2021, ArXiv.

Trajectory Alignment: Understanding the Edge of Stability Phenomenon via Bifurcation Theory

Chulhee Yun, Minhak Song, 2023, 2307.04204.

Tighter Lower Bounds for Shuffling SGD: Random Permutations and Beyond

Chulhee Yun, Jaewook Lee, Jaeyoung Cha, 2023, ICML.

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

Suvrit Sra, Shashank Rajput, Chulhee Yun, 2021, ArXiv.

A Critical View of Global Optimality in Deep Learning pdf

Suvrit Sra, Ali Jadbabaie, Chulhee Yun, 2018, ArXiv.

Linear attention is (maybe) all you need (to understand transformer optimization)

S. Sra, Chulhee Yun, Kwangjun Ahn, 2023, ICLR.

PLASTIC: Improving Input and Label Plasticity for Sample Efficient Reinforcement Learning

Chulhee Yun, J. Choo, Se-Young Yun, 2023, 2306.10711.

Face detection using Local Hybrid Patterns

Donghoon Lee, Chang Dong Yoo, Chulhee Yun, 2015, 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

On the Training Instability of Shuffling SGD with Batch Normalization

S. Sra, Chulhee Yun, David Wu, 2023, ICML.

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

S. Sra, Chulhee Yun, A. Jadbabaie, 2021, COLT.