/flow

Computational framework for reinforcement learning in traffic control

Primary LanguagePythonMIT LicenseMIT

Build Status Docs License

Flow

Flow is a computational framework for deep RL and control experiments for traffic microsimulation.

See our website for more information on the application of Flow to several mixed-autonomy traffic scenarios. Other results and videos are available as well.

More information

Getting involved

We welcome your contributions.

Citing Flow

If you use Flow for academic research, you are highly encouraged to cite our paper:

C. Wu, A. Kreidieh, K. Parvate, E. Vinitsky, A. Bayen, "Flow: Architecture and Benchmarking for Reinforcement Learning in Traffic Control," CoRR, vol. abs/1710.05465, 2017. [Online]. Available: https://arxiv.org/abs/1710.05465

Contributors

Cathy Wu, Eugene Vinitsky, Aboudy Kreidieh, Kanaad Parvate, Nishant Kheterpal, Saleh Albeaik, Kathy Jang, Ananth Kuchibhotla. Alumni contributors include Leah Dickstein and Nathan Mandi. Flow is supported by the Mobile Sensing Lab at UC Berkeley and Amazon AWS Machine Learning research grants.