Chulhee Yun

Implicit Bias of Per-sample Adam on Separable Data: Departure from the Full-batch Regime

Implicit Bias and Loss of Plasticity in Matrix Completion: Depth Promotes Low-Rankness

Scaling Laws of SignSGD in Linear Regression: When Does It Outperform SGD?

Minor First, Major Last: A Depth-Induced Implicit Bias of Sharpness-Aware Minimization

The Cost of Robustness: Tighter Bounds on Parameter Complexity for Robust Memorization in ReLU Nets

Through the River: Understanding the Benefit of Schedule-Free Methods for Language Model Training

Parameter Expanded Stochastic Gradient Markov Chain Monte Carlo

From Linear to Nonlinear: Provable Weak-to-Strong Generalization through Feature Learning

Provable Benefit of Random Permutations over Uniform Sampling in Stochastic Coordinate Descent

Understanding Sharpness Dynamics in NN Training with a Minimalist Example: The Effects of Dataset Difficulty, Depth, Stochasticity, and More

Incremental Gradient Descent with Small Epoch Counts is Surprisingly Slow on Ill-Conditioned Problems

Arithmetic Transformers Can Length-Generalize in Both Operand Length and Count

Convergence and Implicit Bias of Gradient Descent on Continual Linear Classification

Lightweight Dataset Pruning without Full Training via Example Difficulty and Prediction Uncertainty

Does SGD really happen in tiny subspaces?