pat-coady/trpo

Trust Region Policy Optimization with TensorFlow and OpenAI Gym

50
/ 100
Established

This project helps researchers and engineers quickly train robotic agents to perform complex movements and tasks in simulated environments. It takes a simulated robot environment, like a walking humanoid or a robotic arm, and outputs a trained 'brain' that allows the robot to learn how to achieve its goals without extensive manual tuning. It's designed for AI researchers, robotics engineers, and students exploring reinforcement learning for control problems.

361 stars. No commits in the last 6 months.

Use this if you need to efficiently train a variety of robotic agents in simulation, from simple pendulums to complex humanoids, using a robust, automatically tuned algorithm.

Not ideal if you are looking for a tool to control physical robots directly or if your primary focus is on very simple, non-continuous control problems.

robotics-simulation reinforcement-learning motion-planning autonomous-agents AI-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

361

Forks

107

Language

Jupyter Notebook

License

MIT

Last pushed

Jun 02, 2020

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/pat-coady/trpo"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.