Multi-player Multi-armed Bandits with Collision-Dependent Reward Distributions
This addresses a more realistic scenario in applications like cognitive radio, offering incremental improvements over existing methods.
The paper tackles the multi-player multi-armed bandits problem where collisions reduce rewards non-zero, proposing the EC3 algorithm that uses error-correction coding to achieve near-centralized regret, with experiments showing coding choice significantly impacts performance.
We study a new stochastic multi-player multi-armed bandits (MP-MAB) problem, where the reward distribution changes if a collision occurs on the arm. Existing literature always assumes a zero reward for involved players if collision happens, but for applications such as cognitive radio, the more realistic scenario is that collision reduces the mean reward but not necessarily to zero. We focus on the more practical no-sensing setting where players do not perceive collisions directly, and propose the Error-Correction Collision Communication (EC3) algorithm that models implicit communication as a reliable communication over noisy channel problem, for which random coding error exponent is used to establish the optimal regret that no communication protocol can beat. Finally, optimizing the tradeoff between code length and decoding error rate leads to a regret that approaches the centralized MP-MAB regret, which represents a natural lower bound. Experiments with practical error-correction codes on both synthetic and real-world datasets demonstrate the superiority of EC3. In particular, the results show that the choice of coding schemes has a profound impact on the regret performance.