Closed leonard-q closed 2 years ago
After the PPO and ICM, RND are modified, the performance of ICM and RND degrades. The performance of ICM and RND should be recovered through the change of the techniques and hyperparameters.
After the PPO and ICM, RND are modified, the performance of ICM and RND degrades. The performance of ICM and RND should be recovered through the change of the techniques and hyperparameters.