Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
2 files | 3.72 MB
Add to list
  1. Search results
  2. Delay-sensitive local plasticity in e...

Delay-sensitive local plasticity in echo state networks

Stefan-Teodor Iacob (UGent) , Spyridon Chavlis, Panayiota Poirazi and Joni Dambre (UGent)
(2023) 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN. In IEEE International Joint Conference on Neural Networks (IJCNN)
Author
Organization
Abstract
Time delays are inherently present in any physical or biological network. However, the role of delays in echo state networks (ESNs) has only been touched upon. In recent years, the use of local plasticity has been explored in the field of reservoir computing, and specifically in ESNs. In this paper, we investigate the role of distance dependent inter-neuron delays in adaptive reservoirs. We introduce a novel ESN design called adaptive distance-based delay network (ADDN), that combines inter-neuron delays with local synaptic plasticity in the reservoir weights using a delay sensitive version of the Bienenstock-Cooper-Munro (BCM) rule. We show that ADDNs perform better on prediction tasks compared to ESNs, regular distance-based delay networks, and ESNs with conventional BCM connections. We optimized the hyperparameters of ADDNs and each of the baseline models using covariance matrix adaptation evolution strategy (CMA-ES). We prove that with ADDNs, we can evolve a single set of hyperparameters that can generate networks which, after unsupervised adaptation, can obtain good performance on different Mackey-Glass sequences with a range of different time constants. By adapting its reservoir weights to the dynamics of the input data, ADDNs can generalize between versions of the same “class” of tasks.
Keywords
Delay-sensitive BCM, BCM, Local plasticity, Distance-based delays, Echo state networks

Downloads

  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 3.08 MB
  • Download
    DS659 acc.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 641.51 KB

Citation

Please use this url to cite or link to this publication:

MLA
Iacob, Stefan-Teodor, et al. “Delay-Sensitive Local Plasticity in Echo State Networks.” 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, IEEE, 2023, doi:10.1109/ijcnn54540.2023.10191901.
APA
Iacob, S.-T., Chavlis, S., Poirazi, P., & Dambre, J. (2023). Delay-sensitive local plasticity in echo state networks. 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN. Presented at the IJCNN2023, the 1st International Conference in Area of Neural Networks theory, analysis and applications, Queensland, Australia. https://doi.org/10.1109/ijcnn54540.2023.10191901
Chicago author-date
Iacob, Stefan-Teodor, Spyridon Chavlis, Panayiota Poirazi, and Joni Dambre. 2023. “Delay-Sensitive Local Plasticity in Echo State Networks.” In 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN. IEEE. https://doi.org/10.1109/ijcnn54540.2023.10191901.
Chicago author-date (all authors)
Iacob, Stefan-Teodor, Spyridon Chavlis, Panayiota Poirazi, and Joni Dambre. 2023. “Delay-Sensitive Local Plasticity in Echo State Networks.” In 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN. IEEE. doi:10.1109/ijcnn54540.2023.10191901.
Vancouver
1.
Iacob S-T, Chavlis S, Poirazi P, Dambre J. Delay-sensitive local plasticity in echo state networks. In: 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN. IEEE; 2023.
IEEE
[1]
S.-T. Iacob, S. Chavlis, P. Poirazi, and J. Dambre, “Delay-sensitive local plasticity in echo state networks,” in 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, Queensland, Australia, 2023.
@inproceedings{01H7FM7HEXHTVKJ94DBVB4ZKW1,
  abstract     = {{Time delays are inherently present in any physical or biological network. However, the role of delays in echo state networks (ESNs) has only been touched upon. In recent years, the use of local plasticity has been explored in the field of reservoir computing, and specifically in ESNs. In this paper, we investigate the role of distance dependent inter-neuron delays in adaptive reservoirs. We introduce a novel ESN design called adaptive distance-based delay network (ADDN), that combines inter-neuron delays with local synaptic plasticity in the reservoir weights using a delay sensitive version of the Bienenstock-Cooper-Munro (BCM) rule. We show that ADDNs perform better on prediction tasks compared to ESNs, regular distance-based delay networks, and ESNs with conventional BCM connections. We optimized the hyperparameters of ADDNs and each of the baseline models using covariance matrix adaptation evolution strategy (CMA-ES). We prove that with ADDNs, we can evolve a single set of hyperparameters that can generate networks which, after unsupervised adaptation, can obtain good performance on different Mackey-Glass sequences with a range of different time constants. By adapting its reservoir weights to the dynamics of the input data, ADDNs can generalize between versions of the same “class” of tasks.}},
  author       = {{Iacob, Stefan-Teodor and Chavlis, Spyridon and Poirazi, Panayiota and Dambre, Joni}},
  booktitle    = {{2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN}},
  isbn         = {{9781665488686}},
  issn         = {{2161-4393}},
  keywords     = {{Delay-sensitive BCM,BCM,Local plasticity,Distance-based delays,Echo state networks}},
  language     = {{eng}},
  location     = {{Queensland, Australia}},
  pages        = {{8}},
  publisher    = {{IEEE}},
  title        = {{Delay-sensitive local plasticity in echo state networks}},
  url          = {{http://doi.org/10.1109/ijcnn54540.2023.10191901}},
  year         = {{2023}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: