Skip to content

KevinHan1209/FederatedRL

Repository files navigation

Federated and Ensembled Reinforcement Learning on Micro UAVs

Completed as part of my summer REU at Columbia University under the supervision of Dr. Brian Plancher. Serves as a universal platform for ensembled federated deep reinforcement learning with many options for customization. Screenshot

Note:

Due to domain randomization, this platform works with my modified version of stable_baselines3 and gym-pybullet-drones. You will find these packages here: https://github.com/KevinHan1209/drone_simulation https://github.com/KevinHan1209/Sb3_FederatedRL

About

Creating a general platform to use federated reinforcement learning algorithms, compatible with Gym environments and stable_baselines3 algorithms. Main algorithms in server.py and fedsvrpg_m.py

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors