5 papers
Anytime Acceleration of Gradient Descent
Zihan Zhang, Jason D. Lee, Simon S. Du +1
This work investigates stepsize-based acceleration of gradient descent with {\em anytime} convergence guarantees. For smooth (non-strongly) convex optimization, we propose a stepsi…
Horizon-Free Regret for Linear Markov Decision Processes
Zihan Zhang, Jason D. Lee, Yuxin Chen +1
A recent line of works showed regret bounds in reinforcement learning (RL) can be (nearly) independent of planning horizon, a.k.a.~the horizon-free bounds. However, these regret bo…
Optimal Multi-Distribution Learning
Zihan Zhang, Wenhao Zhan, Yuxin Chen +2
Multi-distribution learning (MDL), which seeks to learn a shared model that minimizes the worst-case risk across distinct data distributions, has emerged as a unified framework…
Dichotomy of Early and Late Phase Implicit Biases Can Provably Induce Grokking
Kaifeng Lyu, Jikai Jin, Zhiyuan Li +3
Recent work by Power et al. (2022) highlighted a surprising "grokking" phenomenon in learning arithmetic tasks: a neural net first "memorizes" the training set, resulting in perfec…
Robust Offline Reinforcement Learning -- Certify the Confidence Interval
Jiarui Yao, Simon Shaolei Du
Currently, reinforcement learning (RL), especially deep RL, has received more and more attention in the research area. However, the security of RL has been an obvious problem due t…