<ul><li>A new paper published on arXiv proposes the influential bandit problem, a multi-armed bandit formulation that considers interdependencies and non-stationary environments.</li><li>The problem models arm interactions through an unknown interaction matrix that governs the dynamics of arm losses.</li><li>The paper establishes regret lower bounds for standard bandit algorithms and introduces a new algorithm based on a lower confidence bound (LCB) estimator.</li><li>Empirical evaluations demonstrate the presence of inter-arm influence and confirm the superior performance of the proposed method compared to conventional bandit algorithms.</li></ul>

Influential Bandits: Pulling an Arm May Change the Environment

Discover more