StageLight

• This work corresponds to my Paper "Deep Reinforcement Learning of Traffic Signal Control, a Staged Multiscale Learning Approach".
• Using PyTorch and SUMO (Simulation of Urban MObility) with the TraCI API.
• Converge faster!

StageLight.mp4

---

Rainbow Paper

1. Playing Atari with Deep Reinforcement Learning [arxiv]
2. Deep Reinforcement Learning with Double Q-learning [arxiv]
3. Dueling Network Architectures for Deep Reinforcement Learning [arxiv]
4. Prioritized Experience Replay [arxiv]
5. Noisy Networks for Exploration [arxiv]
6. A Distributional Perspective on Reinforcement Learning [arxiv]
7. Rainbow: Combining Improvements in Deep Reinforcement Learning [arxiv]

Original Code:

• https://github.com/AndreaVidali/Deep-QLearning-Agent-for-Traffic-Signal-Control

Rainbow Code Reference:

• https://github.com/Curt-Park/rainbow-is-all-you-need/blob/master/08.rainbow.ipynb
• https://github.com/deligentfool/dqn_zoo/blob/master/Rainbow%20DQN/rainbow_dqn.py
• https://github.com/higgsfield/RL-Adventure/blob/master/7.rainbow%20dqn.ipynb
• https://github.com/ashutoshtiwari13/DeepQNetwork-Hub/blob/master/Rainbow_DQN.ipynb
• https://github.com/qfettes/DeepRL-Tutorials/blob/master/08.Rainbow.ipynb