Beta Version 1.25.0

Added

  • Added ProximalPolicyOptimizationClip model.

  • Added ReinforcementLearningActorCriticNeuralNetworkBaseModel model.

Changes

  • Refactored the codes for ProximalPolicyOptimization, VanillaPolicyGradient, ActorCritic and AdvantageActorCritic so that it inherits from ReinforcementLearningActorCriticNeuralNetworkBaseModel.

  • Episode updates now runs at the end of final reinforcement at the same episode. Previously it runs before the first reinforcement at the next episode.

This site uses Just the Docs, a documentation theme for Jekyll.