题　目：Adaptively Learning to Rank Items in Online Platforms

嘉　宾：詹若涵，助理教授，香港科技大学

主持人：孙海龙 助理教授 上海交通大学安泰经济与管理学院

时　间：2023年6月28日（周三）14:00-15:30pm

地　点：安泰经济与管理学院A503

内容简介：

We aim to optimize cumulative user engagement in online platforms by learning to adaptively rank candidate items. This problem is formulated within a contextual-bandits framework. Each action corresponds to an item ranking that accounts for user characteristics and position effects. We adjust the predicted user engagement score for each item by an upper confidence bound to balance exploration and exploitation. Our algorithm selects the ranking action that optimizes the sum of these adjusted engagement scores, which we solve efficiently via maximum weight matching. We prove that our algorithm achieves the regret of O(d\sqrt{nKT｝ when ranking K out of n items in a d-dimensional context space over T rounds, assuming user engagement scores adhere to a generalized linear model. This regret alleviates dependence on the action space size that grows factorial with item size. Experiments conducted on both simulated and real-world datasets demonstrate our algorithm outperforms the baseline.  

演讲人简介：

Ruohan Zhan is an assistant professor of Industrial Engineering and Decision Analytics at the Hong Kong University of Science and Technology. Her research develops methods to innovate data-driven decision making using tools from causal inference, statistics, and machine learning, with particular interest in problems from platform operations and economics. Previously, she received her BS in mathematics from Peking University (2017), her MS in statistics and PhD in computational and applied mathematics from Stanford University (2021), where her doctoral research was advised by Susan Athey. She was a postdoc fellow at Stanford Graduate School of Business (2022).  

欢迎广大师生参加！