rcognita.actors

This module contains actors, i.e., entities that directly calculate actions. Actors are inegrated into controllers (agents).

Remarks:

• All vectors are treated as of type [n,]

• All buffers are treated as of type [L, n] where each row is a vector

• Buffers are updated from bottom to top

Classes

Actor(dim_output, dim_input, …[, …])	Class of actors.
ActorCALF(safe_controller, *args[, …])	Actor using Critic As a Lyapunov Function (CALF) to constrain the optimization.
ActorCLF(*args, **kwargs)	ActorCLF is an actor class that aims to optimize the decay of a Control-Lyapunov function (CLF).
ActorMPC(dim_output, dim_input, …[, …])	Model-predictive control (MPC) actor.
ActorProbabilisticEpisodic(dim_output, dim_input)
ActorProbabilisticEpisodicAC(dim_output, …)
ActorRPO(dim_output, dim_input, …[, …])	Running (objective) Plus Optimal (objective) actor.
ActorRQL(dim_output, dim_input, …[, …])	Rollout Q-learning (RQL) actor.
ActorSQL(dim_output, dim_input, …[, …])	Staked Q-learning (SQL) actor.
ActorTabular(dim_world[, predictor, …])	Actor minimizing the sum of the running objective and the optimal (or estimate thereof) objective of the next step.