Efficient Reinforcement Learning using Gaussian Processes

• Tác giả: Marc Peter Deisenroth
• Ngôn ngữ: eng
• ISBN-13: 978-3866445697
• ISBN: KSP/1000019799
• Ký hiệu phân loại: 004 Data processing || Computer science
• Thông tin xuất bản: Karlsruhe : KIT Scientific Publishing, 2010
• Mô tả vật lý: 1 electronic resource (IX, 205 p. p.)
• Bộ sưu tập: Tài liệu truy cập mở
• ID: 230821

Tóm tắt

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.