Act.A3C: Determines which action the algorithm takes

In MartinKies/RLR: Reinforcement Learning with R

Description

Internal function which is called from Worker.A3C.

Usage

Act.A3C(state, model, model.par, algo.par, game.object, worker.number, epsilon)

Arguments

state	Encoded game state as given by state.2.array
model	A model as given by Setup.Neural.Network.A3C or Setup.Neural.Network.A3C.LSTM
model.par	A list as given by Get.Def.Par.Neural.Network.A3C or Get.Def.Par.Neural.Network.A3C.LSTM
game.object	A Game Object (list) as defined by Get.Game.Object.<NAME>.
worker.number	A number to identify the worker.
epsilon	Current Value of epsilon.

MartinKies/RLR documentation built on Dec. 24, 2019, 10:02 p.m.