Highlighted by Joseph Jin-Chuan Tang
From Simple Reinforcement Learning with Tensorflow Part 7: Action-Selection Strategies for Exploration by Arthur Juliani
…each of the approaches using a DQN trained on the CartPole environment available in the OpenAI gym. The Bayesian Dropout and Boltzmann methods proved most helpful, at least in my experiment. I encourage those interested to play around with the hyperparameters, as I am sure better performan…