File size: 1,368 Bytes
099bec8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Meta PyTorch OpenEnv Hackathon  |  Resources & Materials
Resources & Materials
Official hackathon resources for all finalists. Use these to build, train, and iterate.
1. OpenEnv Core
An interface library for RL post-training with environments.
* GitHub: https://github.com/meta-pytorch/OpenEnv
* Docs: https://meta-pytorch.org/OpenEnv/
2. HuggingFace OpenEnv Hub
* Environments: https://huggingface.co/openenv
* Spaces: https://huggingface.co/openenv/spaces
3. Tutorials
Build, run, and train RL environments and training pipelines.
* All Tutorials: https://github.com/meta-pytorch/OpenEnv/tree/main/tutorial
* Training Examples: https://github.com/meta-pytorch/OpenEnv/tree/main/tutorial/examples
* Environment Examples: https://github.com/meta-pytorch/OpenEnv/tree/main/envs
4. YouTube: Building RL Environments
* https://www.youtube.com/watch?v=0airz7BhBiA
* https://www.youtube.com/watch?v=ap4q4sAK4OY
* https://www.youtube.com/watch?v=Jew4lhAiqnw
* https://www.youtube.com/live/kkCNMz0Ptd8?si=JJ7og8x5qc7_Gi0e 
* https://www.youtube.com/watch?v=Jew4lhAiqnw 
5. Research Papers: Reward Engineering
* https://arxiv.org/abs/2408.10215
* https://arxiv.org/abs/2601.19100
Tip: Chat with these papers using LLMs for quick insights on engineering rewards.
Scaler School of Technology  |  scaler.com/school-of-technology/meta-pytorch-hackathon