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...

Full description

Autores:
Tipo de recurso:
Book
Fecha de publicación:
2010
Institución:
Universidad de Bogotá Jorge Tadeo Lozano
Repositorio:
Expeditio: repositorio UTadeo
Idioma:
eng
OAI Identifier:
oai:expeditiorepositorio.utadeo.edu.co:20.500.12010/17578
Acceso en línea:
https://directory.doabooks.org/handle/20.500.12854/45907
http://hdl.handle.net/20.500.12010/17578
Palabra clave:
Autonomous learning
Gaussian processes
Machine learning
Aprendizaje
Aprendizaje experiencial
Aptitud de aprendizaje
Rights
License
Abierto (Texto Completo)
Description
Summary: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.