Reinforcement Learning Using Quantum Boltzmann Machines

By Daniel Crawford, Anna Levit, Navid Ghadermarzy, Jaspreet S. Oberoi, & Pooya Ronagh

We investigate whether quantum annealers with select chip layouts can outperform classical computers in reinforcement learning tasks. We associate a transverse fi eld Ising spin Hamiltonian with a layout of qubits similar to that of a deep Boltzmann machine (DBM) and use simulated quantum annealing (SQA) to numerically simulate quantum sampling from this system. We design a reinforcement learning algorithm in which the set of visible nodes representing the states and actions of an optimal policy are the fi rst and last layers of the deep network. In absence of a transverse eld, our simulations show that DBMs train more e ffectively than restricted Boltzmann machines (RBM) with the same number of weights. Since sampling from Boltzmann distributions of a DBM is not classically feasible, this is evidence of advantage of a non-Turing sampling oracle. We then develop a framework for training the network as a quantum Boltzmann machine (QBM) in the presence of a signifi cant transverse field for reinforcement learning. This further improves the reinforcement learning method using DBMs.

Presented at: Theory of Quantum Computation, Communication and Cryptography TCQ 2017, and Tokyo Institute of Technology, Nanoscience and Quantum Physics Seminar 

RESEARCH PAPER arXiv preprint