Improved Initialization of State-Space Artificial Neural Networks

被引:0
|
作者
Schoukens, Maarten [1 ]
机构
[1] Eindhoven Univ Technol, Dept Elect Engn, NL-5612 AZ Eindhoven, Netherlands
基金
欧盟地平线“2020”;
关键词
NONLINEAR-SYSTEM IDENTIFICATION; MODELS;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The identification of black-box nonlinear state-space models requires a flexible representation of the state and output equation. Artificial neural networks have proven to provide such a representation. However, as in many identification problems, a nonlinear optimization problem needs to be solved to obtain the model parameters (layer weights and biases). A well-thought initialization of these model parameters can often avoid that the nonlinear optimization algorithm converges to a poorly performing local minimum of the considered cost function. This paper introduces an improved initialization approach for nonlinear state-space models represented as a recurrent artificial neural network and emphasizes the importance of including an explicit linear term in the model structure. Some of the neural network weights are initialized starting from a linear approximation of the nonlinear system, while others are initialized using random values or zeros. The effectiveness of the proposed initialization approach over previously proposed methods is illustrated on two benchmark examples.
引用
收藏
页码:1913 / 1918
页数:6
相关论文
共 50 条
  • [1] Improved Initialization for Nonlinear State-Space Modeling
    Marconato, Anna
    Sjoberg, Jonas
    Suykens, Johan A. K.
    Schoukens, Johan
    [J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2014, 63 (04) : 972 - 980
  • [2] Freeway travel time prediction with state-space neural networks - Modeling state-space dynamics with recurrent neural networks
    van Lint, JWC
    Hoogendoorn, SP
    van Zuylen, HJ
    [J]. ADVANCED TRAFFIC MANAGEMENT SYSTEMS FOR FREEWAYS AND TRAFFIC SIGNAL SYSTEMS 2002: HIGHWAY OPERATIONS, CAPACITY, AND TRAFFIC CONTROL, 2002, (1811): : 30 - 39
  • [3] State-Space Representations of Deep Neural Networks
    Hauser, Michael
    Gunn, Sean
    Saab, Samer, Jr.
    Ray, Asok
    [J]. NEURAL COMPUTATION, 2019, 31 (03) : 538 - 554
  • [4] Order estimation in affine state-space neural networks
    Gil, P
    Henriques, J
    Dourado, A
    Duarte-Ramos, H
    [J]. SMCIA/05: PROCEEDINGS OF THE 2005 IEEE MID-SUMMER WORKSHOP ON SOFT COMPUTING IN INDUSTRIAL APPLICATIONS, 2005, : 132 - 137
  • [5] Identification of State-space Linear Parameter-varying Models Using Artificial Neural Networks
    Bao, Yajie
    Velni, Javad Mohammadpour
    Basina, Aditya
    Shahbakhti, Mahdi
    [J]. IFAC PAPERSONLINE, 2020, 53 (02): : 5286 - 5291
  • [6] System identification with state-space recurrent fuzzy neural networks
    Yu, W
    Ferreyra, A
    [J]. 2004 43RD IEEE CONFERENCE ON DECISION AND CONTROL (CDC), VOLS 1-5, 2004, : 5106 - 5111
  • [7] 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
  • [8] On state-space neural networks for systems identification: Stability and complexity
    Gil, P.
    Henriques, J.
    Dourado, A.
    Duarte-Ramos, H.
    [J]. 2006 IEEE CONFERENCE ON CYBERNETICS AND INTELLIGENT SYSTEMS, VOLS 1 AND 2, 2006, : 632 - +
  • [9] A Digital Circuit Design of State-Space Recurrent Neural Networks
    Lin, Che-Wei
    Wang, Jeen-Shing
    [J]. 2008 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC), VOLS 1-6, 2008, : 1837 - 1841
  • [10] Separate Initialization of Dynamics and Nonlinearities in Nonlinear State-Space Models
    Marconato, Anna
    Sjoeberg, Jonas
    Suykens, Johan
    Schoukens, Johan
    [J]. 2012 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE (I2MTC), 2012, : 2104 - 2108