On the sample complexity of actor-critic method for reinforcement learning with function approximation
Harshat Kumar, Alec Koppel, Alejandro Ribeiro
Topics & Concepts
Reinforcement learningTemporal difference learningMarkov decision processBellman equationFunction approximationComputer scienceMathematical optimizationFunction (biology)Stochastic approximationGradient descentDynamic programmingGeneralizationArtificial intelligenceMathematicsArtificial neural networkMarkov processKey (lock)StatisticsEvolutionary biologyMathematical analysisBiologyComputer securityReinforcement Learning in RoboticsAdvanced Bandit Algorithms ResearchAdaptive Dynamic Programming Control