Near-optimal no-regret algorithms for zero-sum
We propose a new no-regret learning algorithm. When used against an adversary, our algorithm achieves average regret that scales as O (1/√T) with the number T of rounds. This regret bound is optimal but not rare, as there are a multitude of learning algorithms with this regret guarantee. However, wh...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Society for Industrial and Applied Mathematics
2012
|
| Online Access: | http://hdl.handle.net/1721.1/73097 https://orcid.org/0000-0002-5451-0490 |