Efficient Reinforcement Learning using Gaussian Processes

• Autore principale: Deisenroth, Marc Peter (auth)
• Natura: Elettronico Capitolo di libro
• Lingua: inglese
• Pubblicazione: KIT Scientific Publishing 2010
• Serie: Karlsruhe Series on Intelligent Sensor-Actuator-Systems / Karlsruher Institut für Technologie, Intelligent Sensor-Actuator-Systems Laboratory
• Soggetti: autonomous learning; Gaussian processes; control; machine learning; Bayesian inference
• Accesso online: 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...