报告时间：2026年4月14日（周二）下午 16:00-17:00
报告地点：苏州大学天赐庄校区（本部）博远楼407
报告人：刘克勤 副教授，西交利物浦大学
报告摘要：
The main challenge in machine learning is the complexity of high-dimensional problems. The way of data sampling itself becomes crucial during the learning evolution process. The tradeoff between exploration and exploitation is well modeled by the Multi-armed Bandit (MAB) problems. In this talk, we present some recent results from my own group on the techniques innovated to solve such high-dimensional reinforcement learning problems. Furthermore, we will illustrate the process of transforming pure math theories (such as algebraic topology) into applied fields for efficient algorithm developments and optimizations.
报告人简介：
Dr. Keqin Liu is an associate professor in the School of Mathematics and Physics at XJTLU. Dr. Liu's research interests include the development of modern math theory for AI technologies, the extension of pure math theory, especially number theory, and the application of pure math theory to solving practical problems in machine learning, e.g., multi-armed bandit in reinforcement learning. Dr. Liu published 40+ academic papers in top international journals and conferences including Management Science and NeurIPS.
邀请人：唐煜