jayLEE0301/dhrl_official

Official code for "DHRL: A Graph-Based Approach for Long-Horizon and Sparse Hierarchical Reinforcement Learning" (NeurIPS 2022 Oral)

35
/ 100
Emerging

This project offers a method for training AI agents to perform complex, multi-step tasks more efficiently. It takes in environment parameters and desired long-term goals, and outputs a trained agent capable of solving challenging problems like intricate robot locomotion or manipulation. AI researchers and robotics engineers working on autonomous systems or simulated agents would find this useful.

No commits in the last 6 months.

Use this if you are developing AI agents for long-horizon tasks where rewards are sparse, and you need faster training and improved performance compared to standard hierarchical reinforcement learning methods.

Not ideal if you are working with simple, short-term tasks or if your reinforcement learning environment does not require a hierarchical approach.

robotics autonomous-systems AI-agent-training complex-task-automation simulated-environments
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

34

Forks

5

Language

Python

License

MIT

Last pushed

Jan 23, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jayLEE0301/dhrl_official"

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