Nonlinear state space learning with EM and neural networks

被引：2

作者：

de Freitas, JF ^{[1
]}

Niranjan, M ^{[1
]}

Gee, AH ^{[1
]}

机构：

[1] Univ Cambridge, Dept Engn, Cambridge CB2 1PZ, England

来源：

NEURAL NETWORKS FOR SIGNAL PROCESSING VIII | 1998年

关键词：

D O I：

10.1109/NNSP.1998.710655

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, we derive the EM algorithm for nonlinear state space models. We show how this algorithm, in conjunction with the well known techniques of Kalman smoothing, can be used for nonlinear system identification. A multi-layer perceptron, whose derivatives are computed by back-propagation, is used to generate the measurements mapping. We found that the method is intrinsically very powerful, simple, elegant and stable. However, it exhibits very slow convergence.

引用

页码：254 / 263

页数：10

共 50 条

[1] Learning State Space Trajectories in Recurrent Neural Networks
Pearlmutter, Barak A.
[J]. NEURAL COMPUTATION, 1989, 1 (02) : 263 - 269
[2] Dynamic learning with the EM algorithm for neural networks
de Freitas, JFG
Niranjan, M
Gee, AH
[J]. JOURNAL OF VLSI SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2000, 26 (1-2): : 119 - 131
[3] NONLINEAR STATE-SPACE GENERALIZATIONS OF GRAPH CONVOLUTIONAL NEURAL NETWORKS
Ruiz, Luana
Gama, Fernando
Ribeiro, Alejandro
Isufi, Elvin
[J]. 2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 5265 - 5269
[4] Dynamic Learning with the EM Algorithm for Neural Networks
J.F.G. de Freitas
M. Niranjan
A.H. Gee
[J]. Journal of VLSI signal processing systems for signal, image and video technology, 2000, 26 : 119 - 131
[5] Design of iterative learning control by the means of state space neural networks
Czajkowski, Andrzej
Patan, Maciej
Patan, Krzysztof
[J]. 2016 3RD CONFERENCE ON CONTROL AND FAULT-TOLERANT SYSTEMS (SYSTOL), 2016, : 293 - 298
[6] State-Space Recurrent Fuzzy Neural Networks for Nonlinear System Identification
Wen Yu
[J]. Neural Processing Letters, 2005, 22 : 391 - 404
[7] State-space recurrent fuzzy neural networks for nonlinear system identification
Yu, W
[J]. NEURAL PROCESSING LETTERS, 2005, 22 (03) : 391 - 404
[8] NEURAL NETWORKS FOR NONLINEAR STATE ESTIMATION
PARISINI, T
ZOPPOLI, R
[J]. INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL, 1994, 4 (02) : 231 - 248
[9] Alternative EM Algorithms for Nonlinear State-space Models
Wahlstrom, Johan
Jalden, Joakim
Skog, Isaac
Handel, Peter
[J]. 2018 21ST INTERNATIONAL CONFERENCE ON INFORMATION FUSION (FUSION), 2018, : 1260 - 1267
[10] Fast α-weighted EM learning for neural networks of module mixtures
Matsuyama, Y
Furukawa, S
Takeda, N
Ikeda, T
[J]. IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE, 1998, : 2306 - 2311

← 1 2 3 4 5 →