Advancesinreinforcementlearningalgorithmshavemadeitpossibletousethemforoptimalcontrolinseveraldifferentindustrialapplications.Withthisbook,youwillapplyReinforcementLearningtoarangeofproblems,fromcomputergamestoautonomousdriving.ThebookstartsbyintroducingyoutoessentialReinforcementLearningconceptssuchasagents,environments,rewards,andadvantagefunctions.Youwillalsomasterthedistinctionsbetweenon-policyandoff-policyalgorithms,aswellasmodel-freeandmodel-basedalgorithms.YouwillalsolearnaboutseveralReinforcementLearningalgorithms,suchasSARSA,DeepQ-Networks(DQN),DeepDeterministicPolicyGradients(DDPG),AsynchronousAdvantageActor-Critic(A3C),TrustRegionPolicyOptimization(TRPO),andProximalPolicyOptimization(PPO).ThebookwillalsoshowyouhowtocodethesealgorithmsinTensorFlowandPythonandapplythemtosolvecomputergamesfromOpenAIGym.Finally,youwillalsolearnhowtotrainacartodriveautonomouslyintheTorcsracingcarsimulator.Bytheendofthebook,youwillbeabletodesign,build,train,andevaluatefeed-forwardneuralnetworksandconvolutionalneuralnetworks.Youwillalsohavemasteredcodingstate-of-the-artalgorithmsandalsotrainingagentsforvariouscontrolproblems.