Antonin RAFFIN

icon indicating a website link https://araffin.github.io/

icon company @DLR-RM icon location Munich icon location Research Engineer in Robotics and Machine Learning, with a focus on Reinforcement Learning.

Results 882 comments of Antonin RAFFIN

RecurrentPPO

Hello, no worry, we can keep it open.

Different results across HPC and PC while training RL using SBX

> However, I did find that results consistently are same within the same machine. Then it might sound like a limitation of Jax, probably similar to [PyTorch](https://docs.pytorch.org/docs/stable/notes/randomness.html)