Author | : Marc Peter Deisenroth |
File Size | : 48,5 Mb |
Publisher | : KIT Scientific Publishing |
Language | : English |
Release Date | : 08 May 2024 |
ISBN | : 9783866445697 |
Pages | : 226 pages |
Efficient Reinforcement Learning Using Gaussian Processes by Marc Peter Deisenroth Book PDF 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.