Machine Learning is increasingly becoming a technology that directly interacts with human users. This is already evident in search engines, recommender systems, and electronic commerce, while other applications are likely to follow in the near future (e.g., autonomous robotics, smart homes, gaming). In this talk, I argue that observed user behavior can be a key source of knowledge for these systems, but that extracting this knowledge requires learning algorithms that explicitly account for human decision making, human motivation, and human abilities. Towards this goal, the talk explores how integrating microeconomic models of human behavior into the learning process leads to new learning models and algorithms that have provable guarantees and that perform robustly in practice. In particular, the talk presents two models of boundedly rational user behavior under which observable actions reliably reveal user preferences, each paired with a learning algorithm for aggregating these preferences and solving the associated exploration/exploitation problems.
(*) Restrictions apply. Some modeling required.
Thorsten Joachims is a Professor in the Department of Computer Science and the Department of Information Science at Cornell University. His research interests center on a synthesis of theory and system building in machine learning, with applications in language technology, information retrieval, and recommendation. His past research focused on support vector machines, text classification, structured output prediction, convex optimization, learning to rank, learning with preferences, and learning from implicit feedback. In 2001, he finished his dissertation advised by Prof. Katharina Morik at the University of Dortmund. From there he also received his Diplom in Computer Science in 1997. Between 2000 and 2001 he worked as a PostDoc at the GMD Institute for Autonomous Intelligent Systems. From 1994 to 1996 he was a visiting scholar with Prof. Tom Mitchell at Carnegie Mellon University.
Host: Calvin McCarter