Re:Evolving Curricula with Regret-Based Environment Design
by minch on Saturday March 05, 2022 @07:04AM
(#30566768)
Thanks! The demo just shows the final agents after training (30K gradient updates). Interesting work re the reward maximizing curricula. I have not seen this before, so thanks for the pointer.
(ps: I co-wrote a short elementary paper on auto curricula design for RL in 2017 [0])
[0]: https://arxiv.org/abs/1703.07853