Learning to Coordinate via Quantum Entanglement in Multi-Agent Reinforcement Learning
John Gardiner, Orlando Romero, Brendan Tivnan, Nicolò Dal Fabbro, George J. Pappas
TL;DR
The paper tackles coordination under partial observability without communication in multi-agent reinforcement learning by leveraging quantum entanglement as a coordination resource. It introduces QuantumSoftmax to differentiably parameterize quantum measurements and a quantum-coordinator/advisor policy architecture that decouples coordination from local decision-making, integrated into a modified MAPPO algorithm. Empirically, the framework learns strategies exhibiting quantum advantage in single-round nonlocal games and in a Dec-POMDP-based multi-router queueing task, outperforming classical shared-randomness baselines in many settings. The work demonstrates a principled, gradient-based approach to incorporating quantum resources into MARL with potential practical impact as quantum hardware matures, and outlines directions for extending the framework to multi-round, imperfect hardware scenarios and more compact policy representations.
Abstract
The inability to communicate poses a major challenge to coordination in multi-agent reinforcement learning (MARL). Prior work has explored correlating local policies via shared randomness, sometimes in the form of a correlation device, as a mechanism to assist in decentralized decision-making. In contrast, this work introduces the first framework for training MARL agents to exploit shared quantum entanglement as a coordination resource, which permits a larger class of communication-free correlated policies than shared randomness alone. This is motivated by well-known results in quantum physics which posit that, for certain single-round cooperative games with no communication, shared quantum entanglement enables strategies that outperform those that only use shared randomness. In such cases, we say that there is quantum advantage. Our framework is based on a novel differentiable policy parameterization that enables optimization over quantum measurements, together with a novel policy architecture that decomposes joint policies into a quantum coordinator and decentralized local actors. To illustrate the effectiveness of our proposed method, we first show that we can learn, purely from experience, strategies that attain quantum advantage in single-round games that are treated as black box oracles. We then demonstrate how our machinery can learn policies with quantum advantage in an illustrative multi-agent sequential decision-making problem formulated as a decentralized partially observable Markov decision process (Dec-POMDP).