Efficient Reinforcement Learning using Gaussian Processes

• Главный автор: Deisenroth, Marc Peter (auth)
• Формат: Электронный ресурс Глава книги
• Язык: английский
• Опубликовано: KIT Scientific Publishing 2010
• Серии: Karlsruhe Series on Intelligent Sensor-Actuator-Systems / Karlsruher Institut für Technologie, Intelligent Sensor-Actuator-Systems Laboratory
• Предметы: autonomous learning; Gaussian processes; control; machine learning; Bayesian inference
• Online-ссылка: DOAB: download the publication; DOAB: description of the publication

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...