Regret Analysis of Sleeping Competing Bandits
Shinnosuke Uba, Yutaro Yamaguchi · 2026
Read Regret Analysis of Sleeping Competing Bandits by Shinnosuke Uba, Yutaro Yamaguchi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-19700.txt.
Description
Uba and Yamaguchi formulate sleeping competing bandits, where players and arms appear and vanish over time. The competing bandits frame mixes online learning with stable matching. Classic models assume everyone is always available, which real settings break. The authors extend the regret definition to this sleeping case. Their algorithm reaches an asymptotic regret bound that matches their lower bound. The match holds when arms clearly outnumber players. That makes the algorithm asymptotically optimal in that regime, they show.
Catalog record
| CREATOR | Shinnosuke Uba, Yutaro Yamaguchi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71975 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.19700