Deep Reinforcement Learning (DRL) paper for Intelligent Robot Control (Robotics Systems)

DRL for continuous action-space control: Deep Deterministic Policy Gradients (DDPG) in PyTorch;

DRL for discrete action-space control: Deep Q-Network or Deep Q-Learning (DQN/DQL) in TensorFlow

Implemented in both PyTorch and TensorFlow: DDPG (model-free actor-critic) and DQL/DQN (Temporal Differencing)

Available on my GitHub: https://github.com/arasdar/RL

The arXiv paper: https://arxiv.org/pdf/2105.13806