Dynamic cluster formation using populations of spiking neurons

Ammar Belatreche, Rakesh Paul

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    2 Citations (Scopus)

    Abstract

    This paper introduces a novel neuro-dynamic system for adaptive online clustering using populations of spiking neurons and spike-timing dependent plasticity (STDP). Real-valued data samples are temporally encoded into spike events, used by biological neurons to encode information and communicate with one another, and clusters are represented by spiking neuron populations of varying size. The number of clusters is unknown a priori and clusters are learned in an online fashion where each data sample is provided only once. The coincidence detection capability of spiking neurons is utilized for data clustering and clusters are dynamically formed. The structure of the spiking neural network is constantly adjusted through adding and pruning of neuron populations. Besides, the number of neurons within each population constantly adapts as new data arrives. STDP is employed to adjust the strength of synaptic connections and enhance the selectivity of each population to its corresponding group of data. Preliminary experiments were carried out on synthetic and selected benchmark datasets to evaluate the performance of the proposed system. Promising results were obtained, which indicate the viability of spike-based population coding for online data clustering.
    LanguageEnglish
    Title of host publicationUnknown Host Publication
    Pages1-6
    Number of pages6
    DOIs
    Publication statusPublished - Jul 2012
    EventThe 2012 IEEE International Joint Conference on Neural Networks (IJCNN), - Brisbane, Australia
    Duration: 1 Jul 2012 → …

    Conference

    ConferenceThe 2012 IEEE International Joint Conference on Neural Networks (IJCNN),
    Period1/07/12 → …

    Fingerprint

    Neurons
    Plasticity
    Dynamical systems
    Neural networks
    Experiments

    Keywords

    • Spiking Neurons
    • Unsupervised Learning
    • Online
    • Clustering
    • Population Coding
    • STDP
    • Spike Response Model

    Cite this

    Belatreche, A., & Paul, R. (2012). Dynamic cluster formation using populations of spiking neurons. In Unknown Host Publication (pp. 1-6) https://doi.org/10.1109/IJCNN.2012.6252532
    Belatreche, Ammar ; Paul, Rakesh. / Dynamic cluster formation using populations of spiking neurons. Unknown Host Publication. 2012. pp. 1-6
    @inproceedings{26e8521b1a86433c8953da44889888a8,
    title = "Dynamic cluster formation using populations of spiking neurons",
    abstract = "This paper introduces a novel neuro-dynamic system for adaptive online clustering using populations of spiking neurons and spike-timing dependent plasticity (STDP). Real-valued data samples are temporally encoded into spike events, used by biological neurons to encode information and communicate with one another, and clusters are represented by spiking neuron populations of varying size. The number of clusters is unknown a priori and clusters are learned in an online fashion where each data sample is provided only once. The coincidence detection capability of spiking neurons is utilized for data clustering and clusters are dynamically formed. The structure of the spiking neural network is constantly adjusted through adding and pruning of neuron populations. Besides, the number of neurons within each population constantly adapts as new data arrives. STDP is employed to adjust the strength of synaptic connections and enhance the selectivity of each population to its corresponding group of data. Preliminary experiments were carried out on synthetic and selected benchmark datasets to evaluate the performance of the proposed system. Promising results were obtained, which indicate the viability of spike-based population coding for online data clustering.",
    keywords = "Spiking Neurons, Unsupervised Learning, Online, Clustering, Population Coding, STDP, Spike Response Model",
    author = "Ammar Belatreche and Rakesh Paul",
    year = "2012",
    month = "7",
    doi = "10.1109/IJCNN.2012.6252532",
    language = "English",
    isbn = "978-1-4673-1488-6",
    pages = "1--6",
    booktitle = "Unknown Host Publication",

    }

    Belatreche, A & Paul, R 2012, Dynamic cluster formation using populations of spiking neurons. in Unknown Host Publication. pp. 1-6, The 2012 IEEE International Joint Conference on Neural Networks (IJCNN), 1/07/12. https://doi.org/10.1109/IJCNN.2012.6252532

    Dynamic cluster formation using populations of spiking neurons. / Belatreche, Ammar; Paul, Rakesh.

    Unknown Host Publication. 2012. p. 1-6.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    TY - GEN

    T1 - Dynamic cluster formation using populations of spiking neurons

    AU - Belatreche, Ammar

    AU - Paul, Rakesh

    PY - 2012/7

    Y1 - 2012/7

    N2 - This paper introduces a novel neuro-dynamic system for adaptive online clustering using populations of spiking neurons and spike-timing dependent plasticity (STDP). Real-valued data samples are temporally encoded into spike events, used by biological neurons to encode information and communicate with one another, and clusters are represented by spiking neuron populations of varying size. The number of clusters is unknown a priori and clusters are learned in an online fashion where each data sample is provided only once. The coincidence detection capability of spiking neurons is utilized for data clustering and clusters are dynamically formed. The structure of the spiking neural network is constantly adjusted through adding and pruning of neuron populations. Besides, the number of neurons within each population constantly adapts as new data arrives. STDP is employed to adjust the strength of synaptic connections and enhance the selectivity of each population to its corresponding group of data. Preliminary experiments were carried out on synthetic and selected benchmark datasets to evaluate the performance of the proposed system. Promising results were obtained, which indicate the viability of spike-based population coding for online data clustering.

    AB - This paper introduces a novel neuro-dynamic system for adaptive online clustering using populations of spiking neurons and spike-timing dependent plasticity (STDP). Real-valued data samples are temporally encoded into spike events, used by biological neurons to encode information and communicate with one another, and clusters are represented by spiking neuron populations of varying size. The number of clusters is unknown a priori and clusters are learned in an online fashion where each data sample is provided only once. The coincidence detection capability of spiking neurons is utilized for data clustering and clusters are dynamically formed. The structure of the spiking neural network is constantly adjusted through adding and pruning of neuron populations. Besides, the number of neurons within each population constantly adapts as new data arrives. STDP is employed to adjust the strength of synaptic connections and enhance the selectivity of each population to its corresponding group of data. Preliminary experiments were carried out on synthetic and selected benchmark datasets to evaluate the performance of the proposed system. Promising results were obtained, which indicate the viability of spike-based population coding for online data clustering.

    KW - Spiking Neurons

    KW - Unsupervised Learning

    KW - Online

    KW - Clustering

    KW - Population Coding

    KW - STDP

    KW - Spike Response Model

    U2 - 10.1109/IJCNN.2012.6252532

    DO - 10.1109/IJCNN.2012.6252532

    M3 - Conference contribution

    SN - 978-1-4673-1488-6

    SP - 1

    EP - 6

    BT - Unknown Host Publication

    ER -