Title: Towards Imprecise Generalisation: From Invariance to Heterogeneity
Date: Tuesday, March 12th, 2024
Time: 10:00 am
Place: E7-5353
Abstract: This talk addresses the challenges of out-of-domain (OOD) generalisation in artificial intelligence and machine learning, highlighting the ambiguity in defining OOD generalisation compared to in-domain (IID) generalisation. The speaker will discuss the concept of imprecise learning, its relation to imprecise probability, and initial work in domain generalisation (DG) problems. The aim is to explore the intersection of learning algorithms and decision-making processes to better understand and advance OOD generalisation.
Speaker Bio: Krikamol Muandet is a chief scientist and tenure-track faculty member at CISPA Helmholtz Center for Information Security, Saarbrücken, Germany. With a Ph.D. in computer science from the University of Tübingen, he has an extensive background in machine learning, having held positions at the Max Planck Institute for Intelligent Systems and Mahidol University. His research interests are broad, with significant contributions to the field of machine learning, and he has served in various editorial roles for major conferences in the field.
This talk is a unique opportunity for those interested in the cutting-edge challenges and advancements in machine learning, especially in the context of generalisation across diverse environments.