DocumentCode :
3117315
Title :
The Neuro Slot Car Racer: Reinforcement Learning in a Real World Setting
Author :
Kietzmann, Tim C. ; Riedmiller, Martin
Author_Institution :
Neuroinformatics Group, Univ. of Osnabruck, Osnabruck, Germany
fYear :
2009
fDate :
13-15 Dec. 2009
Firstpage :
311
Lastpage :
316
Abstract :
This paper describes a novel real-world reinforcement learning application: The Neuro Slot Car Racer. In addition to presenting the system and first results based on Neural Fitted Q-Iteration, a standard batch reinforcement learning technique, an extension is proposed that is capable of improving training times and results by allowing for a reduction of samples required for successful training. The Neuralgic Pattern Selection approach achieves this by applying a failure-probability function which emphasizes neuralgic parts of the state space during sampling.
Keywords :
iterative methods; learning (artificial intelligence); neural nets; probability; failure probability function; neural fitted Q-iteration; neuralgic pattern selection approach; neuro slot car racer; real world setting; reinforcement learning; Application software; Benchmark testing; Cognitive science; Computer science; Control systems; Machine learning; Machine vision; Sampling methods; State-space methods; System testing; offline reinforcement learning; pattern selection; real-world application;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Applications, 2009. ICMLA '09. International Conference on
Conference_Location :
Miami Beach, FL
Print_ISBN :
978-0-7695-3926-3
Type :
conf
DOI :
10.1109/ICMLA.2009.15
Filename :
5381535
Link To Document :
بازگشت