Strategic Arms with Side Communication Prevail Over Low-Regret MAB Algorithms - Archive ouverte HAL
Journal Articles ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Year : 2024

Strategic Arms with Side Communication Prevail Over Low-Regret MAB Algorithms

Abstract

In the strategic multi-armed bandit setting, when arms possess perfect information about the player's behavior, they can establish an equilibrium where: 1. they retain almost all of their value, 2. they leave the player with a substantial (linear) regret. This study illustrates that, even if complete information is not publicly available to all arms but is shared among them, it is possible to achieve a similar equilibrium. The primary challenge lies in designing a communication protocol that incentivizes the arms to communicate truthfully.
Fichier principal
Vignette du fichier
Strategic_Arms_With_Side_Communication.pdf (316.85 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04669024 , version 1 (29-08-2024)

Licence

Copyright

Identifiers

Cite

Ahmed Ben Yahmed, Clément Calauzènes, Vianney Perchet. Strategic Arms with Side Communication Prevail Over Low-Regret MAB Algorithms. ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024, pp.7435-7439. ⟨10.1109/ICASSP48485.2024.10446895⟩. ⟨hal-04669024⟩
82 View
16 Download

Altmetric

Share

More