| Zeit: | 23. April 2024 |
|---|---|
| Download als iCal: |
|
Dr. Celestine Mendler-Dünner
Max Planck Institute for Intelligent Systems & Tübingen AI Center
Tübingen, Germany
Tuesday 2024-04-23 4 p.m.
IST Seminar Room 2.255 - Pfaffenwaldring 9 - Campus Stuttgart-Vaihingen
Abstract
Predictions in the social world generally influence the target of prediction, a phenomenon known as performativity. Self-fulfilling and self-negating predictions are examples of performativity. Of fundamental importance to economics, finance, and the social sciences, the notion has long been absent from the development of machine learning. In machine learning applications, performativity surfaces as distribution shift. A predictive model deployed on a digital platform, for example, influences consumption and thereby changes the data-generating distribution. I will provide an introduction to the recently founded area of performative prediction that provides a definition and conceptual framework to study performativity in machine learning. A consequence of performative prediction is a natural equilibrium notion that corresponds to a fixed point of retraining. Another consequence is a distinction between learning and steering, two mechanisms at play in performative prediction. The notion of steering is in turn intimately related to questions of power in digital markets. I will focus on highlighting the key technical results in performative prediction, measurement and optimization challenges, as well as connections to statistics, game theory, and causality.
Biographical Information
Celestine Mendler-Dünner is a Principal Investigator at the ELLIS Institute Tübingen, co-affiliated with the MPI for Intelligent Systems and the Tübingen AI Center. Her research focuses on machine learning in social context and the role of prediction in digital economies. Celestine obtained her PhD in Computer Science from ETH Zurich in collaboration with IBM Research, Europe. Prior to joining the ELLIS Institute, she spent two years as an SNSF postdoctoral fellow at UC Berkeley, and two years as a group leader at the MPI-IS. For the high industrial impact of her work she was awarded the IBM Research Devision Award, the Fritz Kutter Award and the ETH Medal. She is an ELLIS scholar and a fellow of the Elisabeth-Schiemann-Kolleg, she was program chair for EAAMO’23 and serves as senior area chair for NeurIPS and ICML.