Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/1823
Title: | A hierarchical incremental learning approach to task decomposition |
Authors: | Guan, SU Li, P |
Keywords: | Neural network;Task decomposition;Incremental learning;Ordering |
Issue Date: | 2002 |
Publisher: | Freund & Pettman |
Citation: | Journal of Intelligent Systems. 12 (3) 201-226 |
Abstract: | In this paper, we suggest a new task decomposition method – hierarchical incremental class learning (HICL). In this approach, a -class problem is divided into sub-problems. The sub-problems are learnt sequentially in a hierarchical structure with sub-networks. Each sub-network takes the output from the sub-network immediately below it as well as the original input as its input. The output from each sub-network contains one more class than the sub-network immediately below it, and this output is fed into the sub-network above it. It not only reduces harmful interference among hidden layers, but also facilitates information transfer between classes during training. The later sub-networks can obtain learnt information from the earlier sub-networks. We also proposed two ordering algorithms – MSEF and MSEF-FLD to determine the hierarchical relationship between the sub-networks. The proposed HICL approach shows smaller regression error and classification error than the class decomposition and retraining approaches. |
URI: | http://bura.brunel.ac.uk/handle/2438/1823 |
ISSN: | 0334-1860 |
Appears in Collections: | Electronic and Computer Engineering Dept of Electronic and Electrical Engineering Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
A Hierarchical Incremental Learning Approach to Task Decomposition.txt | 285 B | Text | View/Open |
Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.