devsisters/DQN-tensorflow
TensorFlow implementation of the Deep Q-Network algorithm from the famous DeepMind paper on human-level control through deep reinforcement learning.
Not currently ranked — collecting fresh signals.
star history
This repository implements a Deep Q-Network (DQN) for training agents to play Atari games. It uses TensorFlow to build a convolutional neural network that learns Q-values through experience replay (to decorrelate updates) and fixed target networks (to stabilize training). The implementation supports gym environments and includes variants like Dueling DQN and Double DQN (DDQN).
Frequently asked
- What is devsisters/DQN-tensorflow?
- TensorFlow implementation of the Deep Q-Network algorithm from the famous DeepMind paper on human-level control through deep reinforcement learning.
- Is DQN-tensorflow open source?
- Yes — devsisters/DQN-tensorflow is open source, released under the MIT license.
- What language is DQN-tensorflow written in?
- devsisters/DQN-tensorflow is primarily written in Python.
- How popular is DQN-tensorflow?
- devsisters/DQN-tensorflow has 2.6k stars on GitHub.
- Where can I find DQN-tensorflow?
- devsisters/DQN-tensorflow is on GitHub at https://github.com/devsisters/DQN-tensorflow.