Accelerated Optimal Topology Search for Two-Hidden-Layer Feedforward Neural Networks

被引：4

作者：

Thomas, Alan J. ^{[1
]}

Walters, Simon D. ^{[1
]}

Petridis, Miltos ^{[1
]}

Gheytassi, Saeed Malekshahi ^{[1
]}

Morgan, Robert E. ^{[1
]}

机构：

[1] Univ Brighton, Sch Comp Engn & Math, Brighton, E Sussex, England

来源：

ENGINEERING APPLICATIONS OF NEURAL NETWORKS, EANN 2016 | 2016年 / 629卷

关键词：

Two-hidden-layer feedforward; ANN; Exhaustive search; Optimal topology; Optimal node ratio; Heurix; Universal function approximation; ALGORITHMS; ENGINE;

D O I：

10.1007/978-3-319-44188-7_19

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Two-hidden-layer feedforward neural networks are investigated for the existence of an optimal hidden node ratio. In the experiments, the heuristic n(1) = int(0.5n(h) +1), where n(1) is the number of nodes in the first hidden layer and n(h) is the total number of hidden nodes, found networks with generalisation errors, on average, just 0.023 %-0.056 % greater than those found by exhaustive search. This reduced the complexity of an exhaustive search from quadratic, to linear in nh, with very little penalty. Further reductions in search complexity to logarithmic could be possible using existing methods developed by the Authors.

引用

页码：253 / 266

页数：14

共 50 条

[1] Simplification of a specific two-hidden-layer feedforward networks
Chen, L
Huang, GB
Siew, CK
[J]. ICICS-PCM 2003, VOLS 1-3, PROCEEDINGS, 2003, : 1000 - 1003
[2] Learning capability and storage capacity of two-hidden-layer feedforward networks
Huang, GB
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (02): : 274 - 281
[3] The research on an algorithm for special two-hidden-layer artificial neural networks
Ren, JD
Huang, HY
Bao, R
[J]. 2003 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-5, PROCEEDINGS, 2003, : 1127 - 1131
[4] A new algorithm to design compact two-hidden-layer artificial neural networks
Islam, MM
Murase, K
[J]. NEURAL NETWORKS, 2001, 14 (09) : 1265 - 1278
[5] Hardness of Noise-Free Learning for Two-Hidden-Layer Neural Networks
Chen, Sitan
Gollakota, Aravind
Klivans, Adam R.
Meka, Raghu
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35, NEURIPS 2022, 2022,
[6] Optimal hidden structure for feedforward neural networks
Bachiller, P.
Perez, R. M.
Martinez, P.
Aguilar, P. L.
Diaz, P.
[J]. COMPUTATIONAL INTELLIGENCE: THEORY AND APPLICATIONS, 1999, 1625 : 684 - 685
[7] Approximation capability of two hidden layer feedforward neural networks with fixed weights
Guliyev, Namig J.
Ismailov, Vugar E.
[J]. NEUROCOMPUTING, 2018, 316 : 262 - 269
[8] Feedforward Neural Networks with a Hidden Layer Regularization Method
Alemu, Habtamu Zegeye
Wu, Wei
Zhao, Junhong
[J]. SYMMETRY-BASEL, 2018, 10 (10):
[9] Modular Expansion of the Hidden Layer in Single Layer Feedforward Neural Networks
Tissera, Migel D.
McDonnell, Mark D.
[J]. 2016 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2016, : 2939 - 2945
[10] Classification ability of single hidden layer feedforward neural networks
Huang, GB
Chen, YQ
Babri, HA
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 2000, 11 (03): : 799 - 801

← 1 2 3 4 5 →