Efficient reinforcement learning using Gaussian processes

Autori

• Marc Peter Deisenroth

Parametri

• Computer, IT, programmazione
• Karlsruhe series on intelligent sensor actuator systems

Maggiori informazioni sul libro

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.

Per saperne di più

Lingua: Inglese

Pubblicato: 2010