Florian Bauer

icon company Technical University of Munich icon location Munich, Germany icon location Computer Science student at TUM

Results 2 comments of Florian Bauer

[Feature Request] Group Relative Proximity Optimization (GRPO)

> Although, it won't be the most optimal solution memory-wise, it's possible to compute rewards of different actions from the environment at the same step by cloning the `env` object....

[Feature Request] Implement GRPO

Hi @araffin, thanks for your review. > is indeed a hack that might work in some environments (the one that can be reset to an exact state) and seems to...