monoelh/deep-reinforcement-learning_DDQN_PPO_HER

MLP-framework (pure numpy) and DDQN-framework for OpenAI's Gym games. +test code for PPO added. +Hindsight Experience Replay(HER) bitflip-DQN example. +prioritized replay.

语言：Jupyter Notebook · GitHub 星标：19 · Fork：2

话题：advantage-actor-critic · ddqn · ddqn-framework · deep-q-network · deep-reinforcement-learning · game · hindsight-experience-replay · mlp-framework

在 GitHub 上查看原仓库