信息管理与商业智能系学术讲座

时    间：2023年4月18日（周二）10:00-11:30

地    点：史带楼304室

主持人：信息管理与商业智能系 青年副研究员 刘子博

题    目：Learning Across Bandits in High Dimension via Robust Statistics

主讲人：Kan Xu ,University of Pennsylvania

摘    要：

Decision-makers often face the "many bandits" problem, where one must simultaneously learn across related but heterogeneous contextual bandit instances. For instance, a large retailer may wish to dynamically learn product demand across many stores to solve pricing or inventory problems, making it desirable to learn jointly for stores serving similar customers; alternatively, a hospital network may wish to dynamically learn patient risk across many providers to allocate personalized interventions, making it desirable to learn jointly for hospitals serving similar patient populations. We study the setting where the unknown parameter in each bandit instance can be decomposed into a global parameter plus a sparse instance-specific term. Then, we propose a novel two-stage estimator that exploits this structure in a sample-efficient way by using a combination of robust statistics (to learn across similar instances) and LASSO regression (to debias the results). We embed this estimator within a bandit algorithm, and prove that it improves asymptotic regret bounds in the context dimension d; this improvement is exponential for data-poor instances. Finally, we illustrate the value of our approach on synthetic and real datasets.

简    介：

Kan Xu is a final year PhD student at University of Pennsylvania Department of Economics, and will soon join Arizona State University Carey School of Business as an Assistant Professor in Information System. His research focuses on designing novel machine learning algorithms for data-driven decision making, particularly on multitask learning and bandits, with applications in healthcare, dynamic pricing, natural language processing, etc.

信息管理与商业智能系

2023-4-14