Deep Reinforcement Learning with Regularized Convolutional Neural Fitted Q Iteration

We review the deep reinforcement learning setting, in which an agent receiving high-dimensional input from an environment learns a control policy without supervision using multilayer neural networks. We then extend the Neural Fitted Q Iteration value-based reinforcement learning algorithm (Riedmiller et al) by introducing a novel variation which we call Regularized Convolutional Neural Fitted Q Iteration (RCNFQ) that incorporates convolutional neural networks similarly to the Deep Q Network algorithm (Mnih et al) and dropout regularization to improve generalization performance. Finally, we present an implementation of the algorithm and discuss several extensions.