Quantum Projective Simulation with Hamiltonian Evolution: A study in reinforcement learning

TL;DR

This paper explores a quantum version of Projective Simulation for reinforcement learning, analyzing Hamiltonian-based quantum walks and their potential advantages over classical models in AI decision processes.

Contribution

It introduces a quantum mechanical framework for Projective Simulation, including Hamiltonian-based quantum walks and interacting quantum agents, expanding the scope of AI learning models.

Findings

Quantum Hamiltonians enable quantum walks on clip graphs.

Superposition of clips allows for quantum-enhanced perception.

Interacting quantum agents demonstrate complex behaviors.

Abstract

Projective Simulation was introduced as a novel approach to Artificial Intelligence. It involves a deliberation procedure that consists of a random walk on a graph of clips and allows for the learning agent to project itself into the future before committing to an action. Here we study and analyze a quantum mechanical version in which the random walk is performed by two kinds of Hamiltonians. The first kind is implemented by naively embedding the classical model in a quantum model by turning the clips into qubits. The other allows for storing clips in superpositions of qubits allowing for a potentially purely quantum mechanical learning procedure in which the perception of the environment is purely quantum mechanical but the action is classical. We lastly introduce the concept of interacting projective agents for both the classical and quantum mechanical case.