A New Constructive Algorithm for Architectural and Functional Adaptation of Artificial Neural Networks

• Material Type: 文書・図像類
• Title: A New Constructive Algorithm for Architectural and Functional Adaptation of Artificial Neural Networks
• Author/Editor: ISLAM, Md. Monirul; SATTAR, Md. Abdus; AMIN, Md. Faijul; YAO, Xin; Fellow; IEEE; MURASE, Kazuyuki
• Author Heading: ISLAM, Md. Monirul; SATTAR, Md. Abdus; AMIN, Md. Faijul; YAO, Xin; Fellow; IEEE; MURASE, Kazuyuki
• Publication, Distribution, etc.: Institute of Electrical and Electronics Engineers
• Publication Date: 2009-12
• Publication Date (W3CDTF): 2009-12
• Periodical title: IEEE Transactions on Systems, Man, and Cybernetics Part B
• No. or year of volume/issue: 39 6
• Volume: 39
• Issue: 6
• Pages: 1590-1605
• ISSN (Periodical Title): ISSN : 1083-4419
• Text Language Code: eng
• Subject Heading: Architectural adaptation; artificial neural networks(ANNs); constructive approach; functional adaptation; generalization ability
• Target Audience: 一般
• Note (General): The generalization ability of artificial neural networks (ANNs) is greatly dependent on their architectures. Constructive algorithms provide an attractive automatic way of determining a near-optimal ANN architecture for a given problem. Several such algorithms have been proposed in the literature and shown their effectiveness. This paper presents a new constructive algorithm (NCA) in automatically determining ANN architectures. Unlike most previous studies on detemining ANN architectures, NCA puts emphasis on architectural adaptation and functional adaptation in its architecture determination process. It uses a constructive approach to determine the number of hidden layers in an ANN and of neurons in each hidden layer. To achieve functional adaptation, NCA trains hidden neurons in the ANN by using different training sets that were created by employing a similar concept used in the boosting algorithm. The purpose of using different training sets is to encourage hidden neurons to learn different parts or aspects of the training data so that the ANN can learn the whole training data in a better way. In this paper, the convergence and computational issues of NCA are analytically studied. The computational complexity of NCA is found to be O(W x P_t x T), where W is the number of weights in the ANN, P_t is the number of training examples, and T is the number of training epochs. This complexity has the same order as what the backpropagation learning algorithm requires for training a fixed ANN architecture. A set of eight classification and two approximation benchmark problems was used to evaluate the performance of NCA. The experimental results show that NCA can produce ANN architectures with fewer hidden neurons and better generalization ability compared to existing constructive and nonconstructive algorithms.
• Source: https://u-fukui.repo.nii.ac.jp/?action=repository_action_common_download&item_id=22420&item_no=1&attribute_id=22&file_no=1
• Access Restrictions: 限定公開
• Rights (production): Institute of Electrical and Electronics Engineers
• Data Provider (Database): 国立情報学研究所 : 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）; https://irdb.nii.ac.jp
• Original Data Provider (Database): 福井大学 : 福井大学学術機関リポジトリ