Quantum error correction for the Toric code using Deep reinforcement learning

We implement a quantum error correction algorithm for the Toric code using Deep reinforcement learning. An action-value Q-function encodes the discounted value of moving a defect to a neighboring site on the square grid (the action) depending on the full set of defects on the torus (the syndrome or state). The Q-function is represented by a deep convolutional neural network. Using the translational invariance on the torus allows for viewing each defect from a central perspective which crucially simplifies the Q-function representation independantly of the number of defect pairs.
The training is done using experience replay, where data from the algorithm being played out is stored and used for batch upgrade of the Q-network.
Performance is close to that achieved by the Minimum Weight Perfect Matching algorithm for moderate system sizes and for the uncorrelated noise model.