Shielded AlgorithmsΒΆ
This section covers the PPO variants used with probabilistic shielding. These classes live under masa/prob_shield/ and are designed for the augmented action spaces introduced by the shielding wrappers.
This section covers the PPO variants used with probabilistic shielding. These classes live under masa/prob_shield/ and are designed for the augmented action spaces introduced by the shielding wrappers.