CEC-Agent/CEC

Official Implementation of NeurIPS'23 Paper "Cross-Episodic Curriculum for Transformer Agents"

34
/ 100
Emerging

This project offers a new method called Cross-Episodic Curriculum (CEC) to enhance the performance and learning speed of AI agents. It takes sequences of online learning trials and mixed-quality demonstrations to train Transformer-based agents. The output is a more efficient and generalized AI policy that can perform complex tasks. This tool is for AI researchers and practitioners focused on developing advanced reinforcement learning and imitation learning agents.

No commits in the last 6 months.

Use this if you are developing AI agents for tasks like those in DeepMind Lab or RoboMimic and need to improve their learning efficiency and generalization through structured curriculum learning.

Not ideal if you are looking for a pre-trained, ready-to-deploy agent for specific applications without wanting to engage in the training and evaluation pipeline.

reinforcement-learning imitation-learning AI-agent-training deep-learning-research robotics-simulation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 11 / 25

How are scores calculated?

Stars

31

Forks

4

Language

Python

License

MIT

Last pushed

Oct 12, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/CEC-Agent/CEC"

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