Meet Me at the Arm: The Cooperative Multi-Armed Bandits Problem with Shareable Arms
By: Xinyi Hu, Aldo Pacchiano
Potential Business Impact:
Helps players share resources without knowing how many others use them.
We study the decentralized multi-player multi-armed bandits (MMAB) problem under a no-sensing setting, where each player receives only their own reward and obtains no information about collisions. Each arm has an unknown capacity, and if the number of players pulling an arm exceeds its capacity, all players involved receive zero reward. This setting generalizes the classical unit-capacity model and introduces new challenges in coordination and capacity discovery under severe feedback limitations. We propose A-CAPELLA (Algorithm for Capacity-Aware Parallel Elimination for Learning and Allocation), a decentralized algorithm that achieves logarithmic regret in this generalized regime. Our main contribution is a collaborative hypothesis testing protocol that enables synchronized successive elimination and capacity estimation through carefully structured collision patterns. This represents a provably efficient learning result in decentralized no-sensing MMAB with unknown arm capacities.
Similar Papers
Distributed Algorithms for Multi-Agent Multi-Armed Bandits with Collision
Machine Learning (CS)
Helps players get more rewards without talking.
Decentralized Asynchronous Multi-player Bandits
Machine Learning (CS)
Helps devices share wireless signals without crashing.
Fair Algorithms with Probing for Multi-Agent Multi-Armed Bandits
Machine Learning (CS)
Fairly shares rewards, making systems work better.