Publication | Closed Access
Learning automata with changing number of actions
141
Citations
0
References
1987
Year
Artificial IntelligenceIncremental LearningEngineeringMachine LearningEducationReinforcement Learning (Educational Psychology)Learning ControlLifelong Reinforcement LearningReinforcement SchemeReinforcement Learning (Computer Engineering)Automaton NetworkRobot LearningAutonomous LearningIntelligent ControlAction Model LearningComputer ScienceLinear Reward-inactionDeep Reinforcement LearningAutomated ReasoningAutomaton OperationLearning Automaton
A reinforcement scheme that is based on the linear reward-inaction updating algorithm is presented for a learning automaton whose action set changes from instant to instant. A learning automaton using the algorithm is shown to be both absolutely expedient and ε-optimal. The simulation results verify the ε-optimality of the algorithm. The results can be extended to the design of general nonlinear absolutely expedient learning algorithms.