Note: For the brax environment I reduced the population size from 1024 to 256 and increased the search iterations by the same factor (300 to 1200) in the main run. I am currently running a brax grid search where I used a population size of 256 but with 500 iterations. I will add the results once they are done.
Cartpole-Easy
Cartpole-Hard
MNIST
Brax
Update: Added the brax-ant gridsearch. Very interesting to see that the hyperparameter ranges appear to be fairly task sensitive. The harder brax task appears to be less robust. Also interestingly the same qualitative patterns appeared in the ARS grid search (note: this used a different range).
evosax
Source Code: https://github.com/RobertTLange/evosax/blob/main/evosax/strategies/open_es.pyNote: For the brax environment I reduced the population size from 1024 to 256 and increased the search iterations by the same factor (300 to 1200) in the main run. I am currently running a brax grid search where I used a population size of 256 but with 500 iterations. I will add the results once they are done.
Update: Added the brax-ant gridsearch. Very interesting to see that the hyperparameter ranges appear to be fairly task sensitive. The harder brax task appears to be less robust. Also interestingly the same qualitative patterns appeared in the ARS grid search (note: this used a different range).