Analysis and improvement of policy gradient estimation

DOI10.1016/J.NEUNET.2011.09.005MaRDI QIDQ448295zbMATHOpenAlexDBLPWikidataFDO

Authors Tingting Zhao, Hirotaka Hachiya, Gang Niu, Masashi Sugiyama

Publication date 30 August 2012

Published in Neural Networks (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.neunet.2011.09.005

zbMATH Keywords

variance reductionreinforcement learningpolicy gradientspolicy gradients with parameter-based exploration

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• Efficient sample reuse in policy gradients with parameter-based exploration
• Variance reduction techniques for gradient estimates in reinforcement learning
• Expected policy gradients for reinforcement learning
• Using Gaussian processes for variance reduction in policy gradient algorithms
• Policy gradient in continuous time

Cites work

• scientific article; zbMATH DE number 1702339 (Why is no real title available?)
• scientific article; zbMATH DE number 1983522 (Why is no real title available?)
• scientific article; zbMATH DE number 1753153 (Why is no real title available?)
• scientific article; zbMATH DE number 194544 (Why is no real title available?)
• 10.1162/1532443041827907
• Approximate gradient methods in policy-space optimization of Markov reward processes
• Simple statistical gradient-following algorithms for connectionist reinforcement learning
• Using Expectation-Maximization for Reinforcement Learning
• Variance reduction techniques for gradient estimates in reinforcement learning

Cited in

(15)

• Estimation and approximation bounds for gradient-based reinforcement learning
• Model-based reinforcement learning with dimension reduction
• Model-based policy gradients with parameter-based exploration by least-squares conditional density estimation
• Expected policy gradients for reinforcement learning
• Derivatives of logarithmic stationary distributions for policy gradient reinforcement learning
• Importance sampling techniques for policy optimization
• Policy search for active fault diagnosis with partially observable state
• scientific article; zbMATH DE number 1753153 (Why is no real title available?)
• A unified algorithm framework for mean-variance optimization in discounted Markov decision processes
• Hessian matrix distribution for Bayesian policy gradient reinforcement learning
• An ODE method to prove the geometric convergence of adaptive stochastic algorithms
• Global convergence of policy gradient methods to (almost) locally optimal policies
• Gradient estimation with simultaneous perturbation and compressive sensing
• Efficient sample reuse in policy gradients with parameter-based exploration
• Smoothing policies and safe policy gradients

This page was built for publication: Analysis and improvement of policy gradient estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q448295)