Multi-Distribution Learning, for Robustness, Fairness, and Collaboration

Channel:

Simons Institute for the Theory of Computing

Subscribers:

68,700

Published on August 26, 2022 12:21:44 PM ● Video Link: https://www.youtube.com/watch?v=hL4iqLGsCmE

Duration: 1:27:45

2,392 views

0

Nika Haghtalab (UC Berkeley)
https://simons.berkeley.edu/talks/tbd-460
Data-Driven Decision Processes Boot Camp

Social and real-world considerations such as robustness, fairness, social welfare, and multi-agent tradeoffs have given rise to multi-distribution learning paradigms. In recent years, these paradigms have been studied by several disconnected communities and under different names, including collaborative learning, distributional robustness, and fair federated learning. In this short tutorial, I will highlight the importance of multi-distribution learning paradigms in general, introduce technical tools for addressing them, and discuss how these problems relate to classical and modern consideration in data driven processes.