Constructive algorithm for fully connected cascade feedforward neural networks

被引：41

作者：

Qiao, Junfei ^{[1
,3
]}

Li, Fanjun ^{[1
,2
,3
]}

Han, Honggui ^{[1
,3
]}

Li, Wenjing ^{[1
,3
]}

机构：

[1] Beijing Univ Technol, Coll Elect Informat & Control Engn, Beijing 100124, Peoples R China

[2] Univ Jinan, Sch Math Sci, Shandong 250022, Peoples R China

[3] Beijing Key Lab Computat Intelligence & Intellige, Beijing 100124, Peoples R China

来源：

NEUROCOMPUTING | 2016年 / 182卷

基金：

中国国家自然科学基金; 中国博士后科学基金;

关键词：

Constructive algorithm; Feedforward neural network; Cascade correlation network; Convergence; Orthogonal least squares; EXTREME LEARNING-MACHINE; APPROXIMATION;

D O I：

10.1016/j.neucom.2015.12.003

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper, a novel constructive algorithm, named fast cascade neural network (FCNN), is proposed to design the fully connected cascade feedforward neural network (FCCFNN). First, a modified index, based on the orthogonal least square method, is derived to select new hidden units from candidate pools. Each hidden unit leads to the maximal reduction of the sum of squared errors. Secondly, the input weights and biases of hidden units are randomly generated and remain unchanged during the learning process. The weights, which connect the input and hidden units with the output units, are calculated after all necessary units have been added. Thirdly, the convergence of FCNN is guaranteed in theory. Finally, the performance of FCNN is evaluated on some artificial and real-world benchmark problems. Simulation results show that the proposed FCNN algorithm has better generalization performance and faster learning speed than some existing algorithms. (C) 2015 Elsevier B.V. All rights reserved.

引用

页码：154 / 164

页数：11

共 50 条

[21] Facial expression recognition using constructive feedforward neural networks
Ma, L
Khorasani, K
[J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2004, 34 (03): : 1588 - 1595
[22] A constructive algorithm for fuzzy neural networks
Mascioli, FMF
Martinelli, G
Rizzi, A
[J]. 1997 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I - V: VOL I: PLENARY, EXPERT SUMMARIES, SPECIAL, AUDIO, UNDERWATER ACOUSTICS, VLSI; VOL II: SPEECH PROCESSING; VOL III: SPEECH PROCESSING, DIGITAL SIGNAL PROCESSING; VOL IV: MULTIDIMENSIONAL SIGNAL PROCESSING, NEURAL NETWORKS - VOL V: STATISTICAL SIGNAL AND ARRAY PROCESSING, APPLICATIONS, 1997, : 3193 - 3196
[23] A multiagent-based constructive approach for feedforward neural networks
Lima, CAM
Coelho, ALV
Von Zuben, FJ
[J]. ADVANCES IN INTELLIGENT DATA ANALYSIS V, 2003, 2810 : 462 - 473
[24] A constructive algorithm for wavelet neural networks
Xu, JH
Ho, DWC
[J]. ADVANCES IN NATURAL COMPUTATION, PT 1, PROCEEDINGS, 2005, 3610 : 730 - 739
[25] Facial Expression Recognition Techniques Using Constructive Feedforward Neural Networks and K-Means Algorithm
Ma, Liying
[J]. ADVANCES IN NEURO-INFORMATION PROCESSING, PT II, 2009, 5507 : 711 - 719
[26] Constructive feedforward neural networks using hermite polynomial activation functions
Ma, LY
Khorasani, K
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (04): : 821 - 833
[27] A hybrid training algorithm for feedforward neural networks
Ben Nasr, Mounir
Chtourou, Mohamed
[J]. NEURAL PROCESSING LETTERS, 2006, 24 (02) : 107 - 117
[28] A TRAINING ALGORITHM FOR BINARY FEEDFORWARD NEURAL NETWORKS
GRAY, DL
MICHEL, AN
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (02): : 176 - 194
[29] A novel learning algorithm for feedforward neural networks
Chen, Huawei
Jin, Fan
[J]. ADVANCES IN NEURAL NETWORKS - ISNN 2006, PT 1, 2006, 3971 : 509 - 514
[30] A Hybrid Training Algorithm for Feedforward Neural Networks
Mounir Ben Nasr
Mohamed Chtourou
[J]. Neural Processing Letters, 2006, 24 : 107 - 117

← 1 2 3 4 5 →