Title :
An algorithm for pruning redundant modules in min-ma modular network [min-ma read min-max]
Author :
Lian, Hui-Cheng ; Lu, Bao-Liang
Author_Institution :
Dept. of Comput. Sci. & Eng., Shanghai Jiao Tong Univ., China
fDate :
31 July-4 Aug. 2005
Abstract :
The min-max modular (M3) network is a framework that is capable of solving large-scale pattern classification problems in a parallel way. The M3 network has been successfully applied to several large-scale real-world problems. When a complex problem is decomposed into a number of separable problems, however, the M3 network suffers from its high redundancy of individual modules. This paper proposes an algorithm, called back-searching (BS) algorithm, to prune these redundant modules. The main idea behind the BS algorithm is to use the actual outputs of the trained M3 network associated with training data to find out the redundant modules by means of ´back searching´. In order to ensure the correctness of the algorithm, we prove two propositions theoretically, namely the sufficient proposition and the necessary proposition, and perform simulations on several benchmark and real-world problems. The simulation results indicate that most of all redundant modules can be pruned by our proposed algorithm and the pruned network has the same generalization performance as the original network.
Keywords :
learning (artificial intelligence); minimax techniques; neural nets; pattern classification; search problems; back-searching; large-scale pattern classification; min-max modular network; necessary proposition; redundant module pruning; sufficient proposition; Brain modeling; Computer science; Intelligent networks; Large-scale systems; Machine learning; Machine learning algorithms; Neural networks; Pattern classification; Tagging; Training data;
Conference_Titel :
Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on
Print_ISBN :
0-7803-9048-2
DOI :
10.1109/IJCNN.2005.1556184