norabelrose/classroom

Preference-based reinforcement learning in PyTorch and JAX with a browser-based GUI.

28
/ 100
Experimental

This tool helps researchers and AI practitioners train reinforcement learning agents using human feedback, rather than complex reward functions. You provide demonstrations or comparisons of an agent's behavior through a simple browser interface, and the system uses these preferences to guide the agent's learning. This is ideal for AI researchers or machine learning engineers developing agents for tasks where specifying precise reward signals is difficult.

No commits in the last 6 months.

Use this if you are developing AI agents and find it challenging to define a perfect mathematical reward function, preferring to guide the agent's learning directly with human judgment.

Not ideal if you need a production-ready system for deploying agents today, as this project is still under active development.

AI-training machine-learning-research agent-development human-in-the-loop-AI robotics-control
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

11

Forks

1

Language

Python

License

MIT

Last pushed

May 23, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/norabelrose/classroom"

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