Constant or Logarithmic Regret in Asynchronous Multiplayer Bandits with Limited Communication.
Hugo RichardEtienne BoursierVianney PerchetPublished in: AISTATS (2024)
Keyphrases
- regret bounds
- asynchronous communication
- worst case
- message transmission
- communication bandwidth
- multi armed bandit
- online learning
- linear regression
- communication systems
- multi armed bandits
- lower bound
- stochastic systems
- expert advice
- educational games
- loss function
- communication protocol
- battery power
- confidence bounds
- upper bound