Efficient Reinforcement Learning using Gaussian Processes

This book examines Gaussian processes in both model-based reinforcement learning (RL) and inference in nonlinear dynamic systems.First, we introduce PILCO, a fully Bayesian approach for efficient RL in continuous-valued state and action spaces when no expert knowledge is available. PILCO takes model...

Ful tanımlama

Kaydedildi:
Detaylı Bibliyografya
Yazar: Deisenroth, Marc Peter
Materyal Türü: Online
Dil:İngilizce
Baskı/Yayın Bilgisi: KIT Scientific Publishing 2021
Konular:
Online Erişim:35389
Etiketler: Etiketle
Etiket eklenmemiş, İlk siz ekleyin!
_version_ 1869524441201901568
author Deisenroth, Marc Peter
author_browse Deisenroth, Marc Peter
author_facet Deisenroth, Marc Peter
author_sort Deisenroth, Marc Peter
collection Directory of Open Access Books
description This book examines Gaussian processes in both model-based reinforcement learning (RL) and inference in nonlinear dynamic systems.First, we introduce PILCO, a fully Bayesian approach for efficient RL in continuous-valued state and action spaces when no expert knowledge is available. PILCO takes model uncertainties consistently into account during long-term planning to reduce model bias. Second, we propose principled algorithms for robust filtering and smoothing in GP dynamic systems.
format Online
id doab-20.500.12854ir-45907
institution Directory of Open Access Books
language eng
publishDate 2021
publishDateRange 2021
publishDateSort 2021
publisher KIT Scientific Publishing
publisherStr KIT Scientific Publishing
record_format ojs
spelling doab-20.500.12854ir-459072023-12-20T18:40:48Z Efficient Reinforcement Learning using Gaussian Processes Deisenroth, Marc Peter QA75.5-76.95 autonomous learning Gaussian processes control machine learning Bayesian inference bic Book Industry Communication::U Computing & information technology::UY Computer science This book examines Gaussian processes in both model-based reinforcement learning (RL) and inference in nonlinear dynamic systems.First, we introduce PILCO, a fully Bayesian approach for efficient RL in continuous-valued state and action spaces when no expert knowledge is available. PILCO takes model uncertainties consistently into account during long-term planning to reduce model bias. Second, we propose principled algorithms for robust filtering and smoothing in GP dynamic systems. 2021-02-11T12:10:46Z 2021-02-11T12:10:46Z 2019-07-30 20:02:01 2010 book 35389 18673813 9783866445697 https://directory.doabooks.org/handle/20.500.12854/45907 eng Karlsruhe Series on Intelligent Sensor-Actuator-Systems / Karlsruher Institut für Technologie, Intelligent Sensor-Actuator-Systems Laboratory image/jpeg Attribution-NonCommercial-NoDerivatives 4.0 International https://www.ksp.kit.edu/9783866445697 KIT Scientific Publishing 10.5445/KSP/1000019799 10.5445/KSP/1000019799 68fffc18-8f7b-44fa-ac7e-0b7d7d979bd2 9783866445697 IX, 205 p. open access
spellingShingle QA75.5-76.95
autonomous learning
Gaussian processes
control
machine learning
Bayesian inference
bic Book Industry Communication::U Computing & information technology::UY Computer science
Deisenroth, Marc Peter
Efficient Reinforcement Learning using Gaussian Processes
title Efficient Reinforcement Learning using Gaussian Processes
title_full Efficient Reinforcement Learning using Gaussian Processes
title_fullStr Efficient Reinforcement Learning using Gaussian Processes
title_full_unstemmed Efficient Reinforcement Learning using Gaussian Processes
title_short Efficient Reinforcement Learning using Gaussian Processes
title_sort efficient reinforcement learning using gaussian processes
topic QA75.5-76.95
autonomous learning
Gaussian processes
control
machine learning
Bayesian inference
bic Book Industry Communication::U Computing & information technology::UY Computer science
topic_facet QA75.5-76.95
autonomous learning
Gaussian processes
control
machine learning
Bayesian inference
bic Book Industry Communication::U Computing & information technology::UY Computer science
url 35389
work_keys_str_mv AT deisenrothmarcpeter efficientreinforcementlearningusinggaussianprocesses