Deep Q-learning with TensorFlow

TensorFlow implementation of Deep Reinforcement Learning papers based on deep Q-learning (DQN).

---

Implemented Papers:

[1] Human-Level Control through Deep Reinforcement Learning
[2] Deep Reinforcement Learning with Double Q-learning
[3] Prioritized Experience Replay
[4] Dueling Network Architectures for Deep Reinforcement Learning

---

To Do:

• update to tensorflow 1.+
• add experimental results

---

Requirements:

• Python 2.7x
• TensorFlow
• Openai gym
• numpy
• tqdm

---

Author:

Mahmoud I.Zidan @ZidanMusk