Autonomous reinforcement learning on raw visual input data in a real world application
Authors:
Open Access
- 1 June 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21614393,p. 1-8
Abstract
We propose a learning architecture, that is able to do reinforcement learning based on raw visual input data. In contrast to previous approaches, not only the control policy is learned. In order to be successful, the system must also autonomously learn, how to extract relevant information out of a high-dimensional stream of input information, for which the semantics are not provided to the learning system. We give a first proof-of-concept of this novel learning architecture on a challenging benchmark, namely visual control of a racing slot car. The resulting policy, learned only by success or failure, is hardly beaten by an experienced human player.Keywords
This publication has 21 references indexed in Scilit:
- Deep, Big, Simple Neural Nets for Handwritten Digit RecognitionNeural Computation, 2010
- Reinforcement learning for robot soccerAutonomous Robots, 2009
- Deep belief net learning in a long-range vision system for autonomous off-road driving2008
- Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition2007
- Reducing the Dimensionality of Data with Neural NetworksScience, 2006
- Extremely randomized treesMachine Learning, 2006
- Interactive learning of mappings from visual percepts to actions2005
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Stable Function Approximation in Dynamic Programming1995
- Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in positionBiological Cybernetics, 1980