SAC Target Entropy - Githubissues

UoA-CARES / cares_reinforcement_learning

CARES Reinforcement Learning Package

11 stars 2 forks source link

Closed beardyFace closed 2 weeks ago

beardyFace commented 1 month ago

Target Entropy update for all SAC variants to -np.prod to account for multi-dimensional action spaces