Using trajectory data to improve Bayesian optimization for reinforcement learning

MaRDI QIDQ2933986zbMATH OpenFDO

Authors Aaron Wilson, Alan Fern, Prasad Tadepalli

Publication date 8 December 2014

Full work available at URL http://jmlr.csail.mit.edu/papers/v15/wilson14a.html

zbMATH Keywords

BayesianlearningMarkov decision processMDPoptimizationpolicy search

Mathematics Subject Classification ID

Bayesian problems; characterization of Bayes procedures (62C10) Bayesian inference (62F15) Learning and adaptive systems in artificial intelligence (68T05) Markov and semi-Markov decision processes (90C40)

Recommendations

• Bayesian optimistic Kullback-Leibler exploration
• No-regret Bayesian optimization with unknown hyperparameters
• Bayesian policy gradient and actor-critic algorithms
• Bayesian optimization for policy search via online-offline experimentation
• Bayesian Reinforcement Learning with Exploration

Cited in

(8)

• Constrained Bayesian optimization with noisy experiments
• Bayesian optimistic Kullback-Leibler exploration
• Variance regularization in sequential Bayesian optimization
• scientific article; zbMATH DE number 6982305 (Why is no real title available?)
• Streaming kernel regression with provably adaptive mean, variance, and regularization
• Using simulation to improve sample-efficiency of Bayesian optimization for bipedal robots
• Robust reinforcement learning with Bayesian optimisation and quadrature
• Bayesian optimization for policy search via online-offline experimentation

This page was built for publication: Using trajectory data to improve Bayesian optimization for reinforcement learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2933986)