Qingyun Wu is a Ph.D. candidate in the Department of Computer Science, University of Virginia. Her research focuses on interactive online learning, including bandit algorithms, reinforcement learning, and their applications in real-world problems. Her research has appeared in multiple top-tier venues, including SIGIR, WWW, KDD, and NeurIPS; and her algorithms have been evaluated in several commercial systems in industry (including Yahoo news recommendation and Snapchat lens recommendation). Qingyun received multiple prestigious awards from the University of Virginia for her excellence in research, including the Virginia Engineering Foundation Fellowship and the Graduate Student Award for Outstanding Research. Her recent work on online learning to rank won the Best Paper Award of SIGIR'2019. She was also selected as one of the Rising Stars in EECS 2019.

AG 1, AG 2, AG 3, INET, AG 4, AG 5, SWS, RG1, MMCI   Info

AG Audience

English

60 Minutes

Saarbrücken

Learning is a predominant theme for any intelligent system, humans or machines. Moving beyond the classical paradigm of learning from past experience, e.g., supervised learning from given labels, a learner needs to actively collect exploratory feedback to learn the unknowns. Considerable challenges arise in such a setting, including sample complexity, costly and even outdated feedback.

In this talk, I will introduce our themed efforts on developing solutions to efficiently explore the unknowns and dynamically adjust to the changes through exploratory feedback. Specifically, I will first present our studies in leveraging special problem structures for efficient exploration. Then I will present our work on empowering the learner to detect and adjust to potential changes in the environment adaptively. Besides, I will also highlight the impact our research has generated in top-valued industry applications, including online learning to rank and interactive recommendation.

Danielle Dalton

--email hidden

Yes

Kaiserslautern

G26

KL 111

SWS Space 2 (6312)

passcode not visible

logged in users only