A hybrid classical-quantum approach to speed-up Q-learning

TL;DR

This paper presents a hybrid classical-quantum method that enhances the efficiency of Q-learning by encoding probability distributions on a quantum register, leading to potential quadratic speed-ups in decision-making processes.

Contribution

It introduces a novel quantum routine for encoding probability distributions and demonstrates its application to improve the computational efficiency of Q-learning algorithms.

Findings

Quadratic performance improvement in decision processes

Formal analysis of computational complexity and approximation error

Potential applications beyond reinforcement learning

Abstract

We introduce a classical-quantum hybrid approach to computation, allowing for a quadratic performance improvement in the decision process of a learning agent. In particular, a quantum routine is described, which encodes on a quantum register the probability distributions that drive action choices in a reinforcement learning set-up. This routine can be employed by itself in several other contexts where decisions are driven by probabilities. After introducing the algorithm and formally evaluating its performance, in terms of computational complexity and maximum approximation error, we discuss in detail how to exploit it in the Q-learning context.