Picture for Yudong Chen

Yudong Chen

Faster Fixed-Point Methods for Multichain MDPs

Add code
Jun 26, 2025
Viaarxiv icon

Optimal Single-Policy Sample Complexity and Transient Coverage for Average-Reward Offline RL

Add code
Jun 26, 2025
Viaarxiv icon

Soft Reasoning: Navigating Solution Spaces in Large Language Models through Controlled Embedding Exploration

Add code
May 30, 2025
Viaarxiv icon

RePaViT: Scalable Vision Transformer Acceleration via Structural Reparameterization on Feedforward Network Layers

Add code
May 28, 2025
Viaarxiv icon

A Piecewise Lyapunov Analysis of Sub-quadratic SGD: Applications to Robust and Quantile Regression

Add code
Apr 15, 2025
Viaarxiv icon

Optimally Installing Strict Equilibria

Add code
Mar 05, 2025
Viaarxiv icon

Re-examining Double Descent and Scaling Laws under Norm-based Capacity via Deterministic Equivalence

Add code
Feb 03, 2025
Figure 1 for Re-examining Double Descent and Scaling Laws under Norm-based Capacity via Deterministic Equivalence
Figure 2 for Re-examining Double Descent and Scaling Laws under Norm-based Capacity via Deterministic Equivalence
Figure 3 for Re-examining Double Descent and Scaling Laws under Norm-based Capacity via Deterministic Equivalence
Figure 4 for Re-examining Double Descent and Scaling Laws under Norm-based Capacity via Deterministic Equivalence
Viaarxiv icon

One-step full gradient suffices for low-rank fine-tuning, provably and efficiently

Add code
Feb 03, 2025
Viaarxiv icon

The Limits of Transfer Reinforcement Learning with Latent Low-rank Structure

Add code
Oct 28, 2024
Viaarxiv icon

Two-Timescale Linear Stochastic Approximation: Constant Stepsizes Go a Long Way

Add code
Oct 16, 2024
Viaarxiv icon