Efficient Reinforcement Learning using Gaussian Processes
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...
Main Author: | |
---|---|
Format: | eBook |
Language: | English |
Published: |
KIT Scientific Publishing
2010
|
Series: | Karlsruhe Series on Intelligent Sensor-Actuator-Systems / Karlsruher Institut für Technologie, Intelligent Sensor-Actuator-Systems Laboratory
|
Subjects: | |
Online Access: | |
Collection: | Directory of Open Access Books - Collection details see MPG.ReNa |
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. |
---|---|
Item Description: | Creative Commons (cc), https://creativecommons.org/licenses/by-nc-nd/4.0/ |
Physical Description: | 1 electronic resource (IX, 205 p. p.) |
ISBN: | 1000019799 9783866445697 |