Information-Theoretic, Statistical and Algorithmic Foundations of Reinforcement Learning

International Symposium on Information Theory (ISIT) 2024

Spectral Methods for Data Science: A Statistical Perspective

2023 Summer School on Theoretical Statistics, PKU BICMR & Math

Statistical and Algorithmic Foundations of Reinforcement Learning

JSM 2023, together with Yuting Wei and Yuejie Chi

Non-asymptotic Analysis for Reinforcement Learning [Part 1][Part 2][Part 3]

SIGMETRICS 2023, together with Yuting Wei and Yuejie Chi

Reinforcement Learning: Fundamentals, Algorithms, and Theory

ICASSP 2022, together with Yuting Wei and Yuejie Chi

Statistical and algorithmic foundations of reinforcement learning

ICSA Applied Statistics Symposium 2021, together with Yuting Wei, Yuejie Chi and Zhengyuan Zhou

Nonconvex Optimization for High-Dimensional Signal Estimation: Spectral and Iterative Methods

European Signal Processing Conference (EUSIPCO) 2020

Taming nonconvexity in information science

ITW 2018, together with Yuejie Chi

Recent advances in nonconvex methods for high-dimensional estimation

ICASSP 2018, together with Yuejie Chi and Yue Lu

TRIAD Lecture Series 2019, Georgia Tech

The power of nonconvex optimization in solving random quadratic systems of equations

Random initialization and implicit regularization in nonconvex statistical estimation

Projected power method: an efficient nonconvex algorithm for joint discrete assignment

Spectral methods meets asymmetry: two recent stories

Inference and uncertainty quantification for noisy matrix completion

Inference and uncertainty quantification for low-rank models

Demystifying the efficiency of reinforcement learning: a few recent stories

On the effectiveness of nonconvex optimization in reinforcement learning

Taming nonconvexity in tensor completion: Fast convergence and uncertainty quantification

Taming nonconvexity in statistical and reinforcement learning

Breaking the sample size barrier in reinforcement learning via model-based approaches (a.k.a. plug-in approaches)

Nonconvex optimization meets statistics: a few recent stories

Bridging convex and nonconvex optimization in noisy matrix completion: stability and uncertainty quantification

Stability, nonconvex optimization, and asymmetry in low-rank matrix estimation

Noisy matrix completion: understanding statistical guarantees of convex relaxation via nonconvex optimization

Asymmetry helps: Eigenvalue and eigenvector analyses of asymmetrically perturbed low-rank matrices

Implicit regularization in nonconvex statistical estimation

Spectral method and regularized MLE are both optimal for top-K ranking

The projected power method: a nonconvex algorithm for discrete problems

The projected power method: an efficient algorithm for joint alignment from pairwise differences

Solving random quadratic systems of equations is nearly as easy as solving linear systems

Modern optimization meets physics: recent progress on phase retrieval

Near-optimal joint object matching via convex relaxation