Algorithms for Reinforcement Learning

DOI10.2200/S00268ED1V01Y201005AIM009zbMath1205.68320OpenAlexW4211221179MaRDI QIDQ3588852

Publication date: 10 September 2010

Published in: Synthesis Lectures on Artificial Intelligence and Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.2200/s00268ed1v01y201005aim009

zbMATH Keywords

simulation Markov decision processes stochastic approximation online learning reinforcement learning planning active learning temporal difference learning function approximation least-squares methods Q-learning PAC-learning natural gradient bias-variance tradeoff overfitting stochastic gradient methods policy gradient actor-critic methods

Mathematics Subject Classification ID

Nonnumerical algorithms (68W05) Learning and adaptive systems in artificial intelligence (68T05) Dynamic programming (90C39) Markov and semi-Markov decision processes (90C40) Research exposition (monographs, survey articles) pertaining to computer science (68-02)

Related Items (41)

Least squares policy iteration with instrumental variables vs. direct policy search: comparison against optimal benchmarks using energy storage ⋮ A Two-Timescale Stochastic Algorithm Framework for Bilevel Optimization: Complexity Analysis and Application to Actor-Critic ⋮ A convex optimization approach to dynamic programming in continuous state and action spaces ⋮ Adaptive playouts for online learning of policies during Monte Carlo tree search ⋮ A unified DC programming framework and efficient DCA based approaches for large scale batch reinforcement learning ⋮ Continuous-action planning for discounted infinite-horizon nonlinear optimal control with Lipschitz values ⋮ Closed-form Approximations in Multi-asset Market Making ⋮ Online spatio-temporal matching in stochastic and dynamic domains ⋮ Unnamed Item ⋮ Efficient model-based reinforcement learning for approximate online optimal control ⋮ Computational Benefits of Intermediate Rewards for Goal-Reaching Policy Learning ⋮ Hypervolume indicator and dominance reward based multi-objective Monte-Carlo tree search ⋮ A systematic study on meta-heuristic approaches for solving the graph coloring problem ⋮ On learning and branching: a survey ⋮ Robust adaptive dynamic programming for linear and nonlinear systems: an overview ⋮ Minimax PAC bounds on the sample complexity of reinforcement learning with a generative model ⋮ Optimal activation of halting multi‐armed bandit models ⋮ Formalization of methods for the development of autonomous artificial intelligence systems ⋮ Dynamic treatment regimes: technical challenges and applications ⋮ Deep reinforcement trading with predictable returns ⋮ Model selection in reinforcement learning ⋮ Approximate Q Learning for Controlled Diffusion Processes and Its Near Optimality ⋮ Reinforcement learning algorithms with function approximation: recent advances and applications ⋮ Crowd computing as a cooperation problem: An evolutionary approach ⋮ Asymptotic analysis of value prediction by well-specified and misspecified models ⋮ Abstraction from demonstration for efficient reinforcement learning in high-dimensional domains ⋮ A unified framework for stochastic optimization ⋮ Markov decision processes with sequential sensor measurements ⋮ A Reinforcement Learning Neural Network for Robotic Manipulator Control ⋮ Unnamed Item ⋮ Unnamed Item ⋮ Proximal algorithms and temporal difference methods for solving fixed point problems ⋮ Some recent advances in learning and adaptation for uncertain feedback control systems ⋮ Preference-based reinforcement learning: evolutionary direct policy search using a preference-based racing algorithm ⋮ Bayesian Exploration for Approximate Dynamic Programming ⋮ Finite-Time Performance of Distributed Temporal-Difference Learning with Linear Function Approximation ⋮ On Convergence of Value Iteration for a Class of Total Cost Markov Decision Processes ⋮ Fundamental design principles for reinforcement learning algorithms ⋮ Empirical Q-Value Iteration ⋮ Unnamed Item ⋮ Unnamed Item

Uses Software

R-MAX

This page was built for publication: Algorithms for Reinforcement Learning