FAST LEARNING WITH INCREMENTAL RBF NETWORKS

被引：126

作者：

FRITZKE, B ^{[1
]}

机构：

[1] RUHR UNIV BOCHUM, INST NEUROINFORMAT, D-44780 BOCHUM, GERMANY

来源：

NEURAL PROCESSING LETTERS | 1994年 / 1卷 / 01期

关键词：

D O I：

10.1007/BF02312392

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a new algorithm for the construction of radial basis function (RBF) networks. The method uses accumulated error information to determine where to insert new units. The diameter of the localized units is chosen based on the mutual distances of the units. To have the distance information always available, it is held up-to-date by a Hebbian learning rule adapted from the ''Neural Gas'' algorithm. The new method has several advantages over existing methods and is able to generate small, well-generalizing networks with comparably few sweeps through the training data.

引用

页码：2 / 5

页数：4

共 50 条

[41] Kernel based learning methods:: Regularization networks and RBF networks
Kudová, P
Neruda, R
[J]. DETERMINISTIC AND STATISTICAL METHODS IN MACHINE LEARNING, 2005, 3635 : 124 - 136
[42] Kernel machine for fast and incremental learning of face
Kang, Woo-Sung
Choi, Jin Young
[J]. 2006 SICE-ICASE INTERNATIONAL JOINT CONFERENCE, VOLS 1-13, 2006, : 5377 - +
[43] Fast Cross-Validation for Incremental Learning
Joulani, Pooria
Gyorgy, Andras
Szepesvari, Csaba
[J]. PROCEEDINGS OF THE TWENTY-FOURTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE (IJCAI), 2015, : 3597 - 3604
[44] Fast incremental learning algorithm using Evolutionary Logic Networks for real-value inputs
Park, Myoung Soo
Choi, Jin Young
[J]. 2006 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORK PROCEEDINGS, VOLS 1-10, 2006, : 2047 - +
[45] Topology Learning Embedding: A Fast and Incremental Method for Manifold Learning
Zhu, Tao
Shen, Furao
Zhao, Jinxi
Liang, Yu
[J]. NEURAL INFORMATION PROCESSING, ICONIP 2017, PT I, 2017, 10634 : 43 - 52
[46] Fast incremental learning methods inspired by biological learning behavior
Yamauchi K.
Oohira T.
Omori T.
[J]. Artificial Life and Robotics, 2005, 9 (3) : 128 - 134
[47] Ml-rbf: RBF Neural Networks for Multi-Label Learning
Zhang, Min-Ling
[J]. NEURAL PROCESSING LETTERS, 2009, 29 (02) : 61 - 74
[48] Ml-rbf: RBF Neural Networks for Multi-Label Learning
Min-Ling Zhang
[J]. Neural Processing Letters, 2009, 29 : 61 - 74
[49] Efficient incremental construction of RBF networks using quasi-gradient method
Reiner, Philip
Wilamowski, Bogdan M.
[J]. NEUROCOMPUTING, 2015, 150 : 349 - 356
[50] A fast training algorithm for RBF networks based on subtractive clustering
Sarimveis, H
Alexandridis, A
Bafas, G
[J]. NEUROCOMPUTING, 2003, 51 : 501 - 505

← 1 2 3 4 5 →