Abstract
Approachability has become a central tool in the analysis of repeated games and online learning. A player plays a repeated vector-valued game against Nature and her objective is to have her long-term average reward inside some target set. The celebrated results of Blackwell provide a 1/√n convergence rate of the expected point-to-set distance if this is achievable, i.e., if the set is approachable. In this paper we provide a characterization for the convergence rates of approachability and show that in some cases a set can be approached with a 1/n rate. Our characterization is solely based on a combination of geometric properties of the set with properties of the repeated game, and not on additional restrictive assumptions on Nature's behavior.
| Original language | English |
|---|---|
| Pages (from-to) | 474-477 |
| Number of pages | 4 |
| Journal | Journal of Machine Learning Research |
| Volume | 30 |
| Publication status | Published - 1 Jan 2013 |
| Externally published | Yes |
| Event | 26th Conference on Learning Theory, COLT 2013 - Princeton, NJ, United States Duration: 12 Jun 2013 → 14 Jun 2013 |