7月12日 Chuan He博士学术报告

发布时间:2025-07-07

报告题目：Complexity of normalized stochastic first-order methods with momentum under heavy-tailed noise

主讲人：Dr. Chuan He, Linköping University 

报告时间：2025年7月12日（周六），9:00-10:00

报告地点：计算机学院A501

主持人：王晓教授

摘要：

In this work, we propose practical normalized stochastic first-order methods with Polyak momentum, multi-extrapolated momentum, and recursive momentum for solving unconstrained optimization problems. These methods employ dynamically updated algorithmic parameters and do not require explicit knowledge of problem-dependent quantities such as the Lipschitz constant or noise bound. We establish first-order oracle complexity results for finding approximate stochastic stationary points under heavy-tailed noise and weakly average smoothness conditions—both of which are weaker than the commonly used bounded variance and mean-squared smoothness assumptions. Our complexity bounds either improve upon or match the best-known results in the literature. Numerical experiments are presented to demonstrate the practical effectiveness of the proposed methods. The paper is available on arXiv at https://arxiv.org/pdf/2506.11214

主讲人简介：

Dr. Chuan He is an Assistant Professor in the Department of Mathematics and also affiliated with WASP at Linköping University. He has published numerous papers in journals such as SIOPT, MOR, JMLR, TMLR, IJOC, and COAP. He was a postdoctoral associate in the Department of Computer Science and Engineering at Universitof Minnesota, working with Professor Ju Sun. He earned his PhD from the Department of Industrial and Systems Engineering at University of Minnesota under the supervision of Professor Zhaosong Lu. He received his bachelor's degree from the School of Mathematical Sciences at Xiamen University, with thesis advised by Professor Wen Huang. His research interests center around data science and optimization, with topics including deep learning, decentralized optimization, large-scale optimization, and high-order methods. He is also interested in applications of machine learning in healthcare, scientific computing, image science, and engineering.