Efficient Reinforcement Learning using Gaussian Processes
Materialtyp:
ArtikelSerie: Utgivningsinformation: KIT Scientific Publishing 2010Beskrivning: 1 electronic resource (IX, 205 p. p.)Innehållstyp: - text
- computer
- online resource
- 9783866445697
Inga fysiska exemplar för denna post
Open Access Unrestricted online access star
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.
Creative Commons Licence cc by-nc-nd cc https://creativecommons.org/licenses/by-nc-nd/4.0/
eng
Freely available e-book