Hierarchical Architecture with Modular Network SOM and Modular Reinforcement Learning

Authors:
Masumi Ishikawa;Kosuke Ueno
Affiliations:
Department of Brain Science and Engineering, Kyushu Institute of Technology, Kitakyushu, Japan 808-0196;Department of Brain Science and Engineering, Kyushu Institute of Technology, Kitakyushu, Japan 808-0196
Venue:
ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part I
Year:
2009

Citing 6
Cited 0

Multiple paired forward and inverse models for motor control

Neural Networks - Special issue on neural control and robotics: biology and technology
Multiple model-based reinforcement learning

Neural Computation
Novelty detection: a review—part 1: statistical approaches

Signal Processing
Task segmentation in a mobile robot by mnSOM: a new approach to training expert modules

Neural Computing and Applications
Brain-Inspired Emergence of Behaviors in Mobile Robots by Reinforcement Learning with Internal Rewards

HIS '08 Proceedings of the 2008 8th International Conference on Hybrid Intelligent Systems
Modular network SOM

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

We propose a hierarchical architecture composed of a modular network SOM (mnSOM) layer and a modular reinforcement learning (mRL) layer. The mnSOM layer models characteristics of a target system, and the mRL layer provides control signals to the target system. Given a set of inputs and outputs from the target system, a winner module which minimizes the mean square output error is determined in the mnSOM layer. The corresponding module in the mRL layer is trained by reinforcement learning to maximize accumulated future rewards. An essential point, here, is that neighborhood learning is adopted at both layers, which guarantees a topology preserving map based on similarity between modules. Its application to a pursuit-evasion game demonstrates usefulness of interpolated modules in providing appropriate control signals. A modular approach to both modeling and control proposed in the paper provides a promising framework for wide-ranging tasks.