Prof. Masashi Sugiyama Talk (Importance Weighting for Adaptive Learning under Distribution Shifts)
Masashi Sugiyama
Abstract
A common assumption in machine learning is that training and test data are drawn from the same underlying distribution. In practice, however, this assumption is frequently violated due to evolving environments, data collection biases, or changes in user behavior. Such distribution shifts present a significant challenge to building reliable and robust machine learning systems. In this talk, I will first review the fundamentals of importance weighting and its role in addressing distribution mismatch. I will then discuss its applications in the post-training of large language models. Finally, I will explore more advanced topics, including sequential distribution shifts and joint distribution shifts.
Speaker
Masashi Sugiyama
Masashi Sugiyama is Director of the RIKEN Center for Advanced Intelligence Project and Professor of Complexity Science and Engineering at the University of Tokyo. His research interests include the theory, algorithms, and applications of machine learning. He has written several books on machine learning, including Density Ratio Estimation in Machine Learning (Cambridge, 2012). He served as program co-chair and general co-chair of the NIPS conference in 2015 and 2016, respectively, and received the Japan Academy Medal in 2017.
Video
Chat is not available.
Successful Page Load