Reconstruction of neuromorphic dynamics from a single scalar time series using variational autoencoder and neural network map

被引:0
|
作者
Kuptsov, Pavel V. [1 ]
Stankevich, Nataliya V.
机构
[1] Russian Acad Sci, Kotelnikov Inst RadioEngn & Elect, Saratov Branch, 38 Zelenaya Str, Saratov 410019, Russia
基金
俄罗斯科学基金会;
关键词
Artificial Neural Network; Dynamical system; Dynamical modeling; Hodgkin-Huxley-type model; Reconstruction dynamics; REPRESENTATION; VARIABLES;
D O I
10.1016/j.chaos.2024.115818
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
This paper examines the reconstruction of a family of dynamical systems with neuromorphic behavior using a single scalar time series. A model of a physiological neuron based on the Hodgkin-Huxley formalism is considered. Single time series of one of its variables is shown to be enough to train a neural network that can operate as a discrete time dynamical system with one control parameter. The neural network system is created in two steps. First, the delay-coordinate embedding vectors are constructed form the original time series and their dimension is reduced with by means of a variational autoencoder to obtain the recovered state-space vectors. It is shown that an appropriate reduced dimension can be determined by analyzing the autoencoder training process. Second, pairs of the recovered state-space vectors at consecutive time steps supplied with a constant value playing the role of a control parameter are used to train another neural network to make it operate as a recurrent map. The regimes of thus created neural network system observed when its control parameter is varied are in very good accordance with those of the original system, though they were not explicitly presented during training.
引用
收藏
页数:12
相关论文
共 50 条
  • [41] DESIGN AND DEVELOPMENT OF A SIT-TO-STAND DEVICE USING A VARIATIONAL AUTOENCODER-BASED DEEP NEURAL NETWORK
    Lyu, Zhijie
    Purwar, Anurag
    PROCEEDINGS OF ASME 2022 INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, IDETC-CIE2022, VOL 7, 2022,
  • [42] Time series prediction using wavelet process neural network
    Ding Gang
    Zhong Shi-Sheng
    Li Yang
    CHINESE PHYSICS B, 2008, 17 (06) : 1998 - 2003
  • [43] Predicting a chaotic time series using a fuzzy neural network
    Maguire, LP
    Roche, B
    McGinnity, TM
    McDaid, LJ
    INFORMATION SCIENCES, 1998, 112 (1-4) : 125 - 136
  • [44] Financial Time Series Forecast Using Neural Network Ensembles
    Tarsauliya, Anupam
    Kala, Rahul
    Tiwari, Ritu
    Shukla, Anupam
    ADVANCES IN SWARM INTELLIGENCE, PT I, 2011, 6728 : 480 - 488
  • [45] Time series analysis using hierarchical neural network classifiers
    Dolenko, SA
    Orlov, YV
    Persiantsev, IG
    Shugai, JS
    PROCEEDINGS OF THE FIFTH JOINT CONFERENCE ON INFORMATION SCIENCES, VOLS 1 AND 2, 2000, : 908 - 911
  • [46] A Synthetic Time-Series Generation Using a Variational Recurrent Autoencoder with an Attention Mechanism in an Industrial Control System
    Jeon, Seungho
    Seo, Jung Taek
    SENSORS, 2024, 24 (01)
  • [47] Landsat Time Series Reconstruction Using a Closed-Form Continuous Neural Network in the Canadian Prairies Region
    Ataabadi, Masoud Babadi
    Pouliot, Darren
    Chen, Dongmei
    Oluwadare, Temitope Seun
    SENSORS, 2025, 25 (05)
  • [48] Radar Image Reconstruction from Raw ADC Data using Parametric Variational Autoencoder with Domain Adaptation
    Stephan, Michael
    Stadelmayer, Thomas
    Santra, Avik
    Fischer, Georg
    Weigel, Robert
    Lurz, Fabian
    2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2021, : 9529 - 9536
  • [49] Anomaly-Based Intrusion Detection From Network Flow Features Using Variational Autoencoder
    Zavrak, Sultan
    Iskefiyeli, Murat
    IEEE ACCESS, 2020, 8 : 108346 - 108358
  • [50] Detecting Web Attacks From HTTP Weblogs Using Variational LSTM Autoencoder Deviation Network
    Jagat, Rikhi Ram
    Sisodia, Dilip Singh
    Singh, Pradeep
    IEEE TRANSACTIONS ON SERVICES COMPUTING, 2024, 17 (05) : 2210 - 2222