GINO-Q: Learning an Asymptotically Optimal Index Policy for Restless Multi-armed Bandits

TL;DR

GINO-Q is a novel algorithm that efficiently learns near-optimal index policies for large-scale restless multi-armed bandits without requiring indexability, overcoming computational challenges and outperforming existing methods.

Contribution

It introduces GINO-Q, a three-timescale stochastic approximation algorithm that decomposes RMABs into manageable subproblems, enabling scalable and flexible policy learning.

Findings

GINO-Q converges faster than existing algorithms.

It performs well on non-indexable RMABs.

It achieves near-optimal policies in large-scale settings.

Abstract

The restless multi-armed bandit (RMAB) framework is a popular model with applications across a wide variety of fields. However, its solution is hindered by the exponentially growing state space (with respect to the number of arms) and the combinatorial action space, making traditional reinforcement learning methods infeasible for large-scale instances. In this paper, we propose GINO-Q, a three-timescale stochastic approximation algorithm designed to learn an asymptotically optimal index policy for RMABs. GINO-Q mitigates the curse of dimensionality by decomposing the RMAB into a series of subproblems, each with the same dimension as a single arm, ensuring that complexity increases linearly with the number of arms. Unlike recently developed Whittle-index-based algorithms, GINO-Q does not require RMABs to be indexable, enhancing its flexibility and applicability. Our experimental results demonstrate that GINO-Q consistently learns near-optimal policies, even for non-indexable RMABs where Whittle-index-based algorithms perform poorly, and it converges significantly faster than existing baselines.