An evolutionary building algorithm for Deep Neural Networks

被引：0

作者：

Zemouri, Ryad ^{[1
]}

机构：

[1] Conservatoire Natl Arts & Metiers, Cedric Lab, 292 Rue St Martin, F-75141 Paris 03, France

来源：

2017 12TH INTERNATIONAL WORKSHOP ON SELF-ORGANIZING MAPS AND LEARNING VECTOR QUANTIZATION, CLUSTERING AND DATA VISUALIZATION (WSOM) | 2017年

关键词：

Deep Neural Network; Deep Learning; Constructive Neural Networks; ADAPTIVE LEARNING RATE; REPRESENTATION; CONVERGENCE;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The increase of the computer power has contributed significantly to the development of the Deep Neural Networks. However, the training phase is more difficult since there are many hidden layers with many connections. The aim of this paper is to improve the learning procedure for Deep Neural Networks. A new method for building an evolutionary DNN is presented. With our method, the user does not have to arbitrary specify the number of hidden layers nor the number of neurons per layer. Illustrative examples are provided to support the theoretical analysis.

引用

页码：21 / 27

页数：7

共 19 条

[1]

[Anonymous], 2011, 2 INT C CONTR INSTR

[2]

[Anonymous], 2013, ICML

[3]

[Anonymous], 2012, MACH LEARN

[4] On adaptive learning rate that guarantees convergence in feedforward networks [J].

Behera, Laxmidhar ;

Kumar, Swagat ;

Patnaik, Awhan .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2006, 17 (05) :1116-1125

[5] Representation Learning: A Review and New Perspectives [J].

Bengio, Yoshua ;

Courville, Aaron ;

Vincent, Pascal .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (08) :1798-1828

[6] Improving scalability of ART neural networks [J].

Benites, Fernando ;

Sapozhnikova, Elena .

NEUROCOMPUTING, 2017, 230 :219-229

[7] Will the real Iris data please stand up? [J].

Bezdek, JC ;

Keller, JM ;

Krishnapuram, R ;

Kuncheva, LI ;

Pal, NR .

IEEE TRANSACTIONS ON FUZZY SYSTEMS, 1999, 7 (03) :368-369

[8] Deep learning with adaptive learning rate using laplacian score [J].

Chandra, B. ;

Sharma, Rajesh K. .

EXPERT SYSTEMS WITH APPLICATIONS, 2016, 63 :1-7

[9]

DUFFNER S, 2007, ONL BACKPR ALG VAL, V4668, P249

[10]

Grossberg S, 2012, NEURAL NETWORKS, V27, P1, DOI [10.1016/j.neunet.2012.09.017, 10.1016/j.neunet.2011.10.011]

← 1 2 →