Multilayer feedforward networks with adaptive spline activation function

被引：81

作者：

Guarnieri, S ^{[1
]}

Piazza, F ^{[1
]}

Uncini, A ^{[1
]}

机构：

[1] Univ Ancona, Dipartimento Elettron & Automat, I-60131 Ancona, Italy

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 1999年 / 10卷 / 03期

关键词：

adaptive activation functions; function shape autotuning; generalization; generalized sigmoidal functions; multilayer perceptron; neural networks; spline neural networks;

D O I：

10.1109/72.761726

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 new adaptive spline activation function neural network (ASNN) is presented. Due to the ASNN's high representation capabilities, networks with a small number of interconnections can be trained to solve both pattern recognition and data processing real-time problems. The main idea is to use a Catmull-Rom cubic spline as the neuron's activation function, which ensures a simple structure suitable for both software and hardware implementation. Experimental results demonstrate improvements in terms of generalization capability and of learning speed in both pattern recognition and data processing tasks.

引用

页码：672 / 683

页数：12

共 50 条

[1] An adaptive activation function for multilayer feedforward neural networks
Yu, CC
Tang, YC
Liu, BD
[J]. 2002 IEEE REGION 10 CONFERENCE ON COMPUTERS, COMMUNICATIONS, CONTROL AND POWER ENGINEERING, VOLS I-III, PROCEEDINGS, 2002, : 645 - 650
[2] A robust learning algorithm for feedforward neural networks with adaptive spline activation function
Hu, LY
Sun, ZQ
[J]. ADVANCES IN NEURAL NETWORKS - ISNN 2005, PT 1, PROCEEDINGS, 2005, 3496 : 566 - 571
[3] Neural networks with adaptive spline activation function
Campolucci, P
Capparelli, F
Guarnieri, S
Piazza, F
Uncini, A
[J]. MELECON '96 - 8TH MEDITERRANEAN ELECTROTECHNICAL CONFERENCE, PROCEEDINGS, VOLS I-III: INDUSTRIAL APPLICATIONS IN POWER SYSTEMS, COMPUTER SCIENCE AND TELECOMMUNICATIONS, 1996, : 1442 - 1445
[4] MULTILAYER FEEDFORWARD NETWORKS WITH A NONPOLYNOMIAL ACTIVATION FUNCTION CAN APPROXIMATE ANY FUNCTION
LESHNO, M
LIN, VY
PINKUS, A
SCHOCKEN, S
[J]. NEURAL NETWORKS, 1993, 6 (06) : 861 - 867
[5] Type-2 fuzzy activation function for multilayer feedforward neural networks
Karaköse, M
Akin, E
[J]. 2004 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN & CYBERNETICS, VOLS 1-7, 2004, : 3762 - 3767
[6] A note on activation function in multilayer feedforward learning
Kamruzzaman, J
Aziz, SM
[J]. PROCEEDING OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-3, 2002, : 519 - 523
[7] Learning and approximation capabilities of adaptive spline activation function neural networks
Vecci, L
Piazza, F
Uncini, A
[J]. NEURAL NETWORKS, 1998, 11 (02) : 259 - 270
[8] Adaptive regression estimation with multilayer feedforward neural networks
Kohler, M
Krzyzak, A
[J]. 2004 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS, 2004, : 467 - 467
[9] Adaptive regression estimation with multilayer feedforward neural networks
Kohler, M
Krzyzak, A
[J]. JOURNAL OF NONPARAMETRIC STATISTICS, 2005, 17 (08) : 891 - 913
[10] Bi-modal derivative adaptive activation function sigmoidal feedforward artificial neural networks
Mishra, Akash
Chandra, Pravin
Ghose, Udayan
Sodhi, Sartaj Singh
[J]. APPLIED SOFT COMPUTING, 2017, 61 : 983 - 994

← 1 2 3 4 5 →