Speaker |
Alexis Bellot |
|---|---|
Affiliation |
DeepMind |
Date |
Friday, 01 November 2024 |
Time |
12:00-13:00 |
Location |
Function Space, UCL Centre for Artificial Intelligence, 1st Floor, 90 High Holborn, London WC1V 6BH |
Link |
https://ucl.zoom.us/j/97245943682 |
Event series |
Jump Trading/ELLIS CSML Seminar Series |
Abstract |
A fundamental task in AI is providing performance guarantees for predictions made in unseen domains. In practice, there can be substantial uncertainty about the distribution of new data, and corresponding variability in the performance of existing predictors. For example, a risk prediction tool fine-tuned on a patient population (e.g. particular hospital, geographic location) may not be equally optimal if deployed on a different patient population that may differ in several aspects. This talk studies this problem through the lens of partial transportability, that combines data from source domains and assumptions about the data generating mechanisms, encoded in causal diagrams, to provide a guarantee on out-of-distribution performance of classification models. We will show that one may consistently predict the worst-case performance of existing classification models, and that, further, one may train classification models to explicitly optimize for worst-case performance in a target domain, under our assumptions. Both these methods may be parameterized with expressive neural networks and implemented with gradient-based optimization schemes. With these results, we hope to provide a fresh perspective on the problem of transfer learning and domain generalization in machine learning. |
Biography |
Alexis Bellot is a research scientist at Google DeepMind in London, UK. He was previously a postdoctoral scholar at Columbia University working with Professor Elias Bareinboim. Prior to Columbia, he graduated with a PhD in Applied Mathematics from the University of Cambridge under the supervision of Professor Mihaela van der Schaar. Alexis works on the study of causality from data and its applications, with an emphasis on methods and theory that combine causality and machine learning to help guarantee the safety and alignment of AI systems. |