Szczegóły publikacji
Opis bibliograficzny
Towards analog implementation of spiking neural networks for audio signals / Maciej WIELGOSZ, Andrzej SKOCZEŃ, Jerzy Dąbrowski, Aleksandra Dąbrowska, Waldemar Tabaczyński // W: Intelligent Computing : proceedings of the 2023 Computing Conference : [22–23 June 2023, London, UK], Vol. 2 / ed. Kohei Arai. — Cham : Springer Nature Switzerland, cop. 2023. — (Lecture Notes in Networks and Systems ; ISSN 2367-3370 ; LNNS 739). — ISBN: 978-3-031-37962-8; e-ISBN: 978-3-031-37963-5. — S. 905–922. — Bibliogr., Abstr. — Publikacja dostępna online od: 2023-08-20. — M. Wielgosz, A. Skoczeń - dod. afiliacja: Jagiellonian University
Autorzy (5)
- AGHWielgosz Maciej
- AGHSkoczeń Andrzej
- Dąbrowski Jerzy
- Dąbrowska Aleksandra
- Tabaczyński Waldemar
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 151053 |
---|---|
Data dodania do BaDAP | 2024-01-08 |
DOI | 10.1007/978-3-031-37963-5_63 |
Rok publikacji | 2023 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Wydawca | Springer |
Czasopismo/seria | Lecture Notes in Networks and Systems |
Abstract
This publication presents a novel approach to the training and deploying Spiking Neural Networks on analog hardware platforms. We proposed a scheme composed of steps involving architectural consideration, training, and mapping protocol. The method employs Genetic Algorithms and the Gradient-based method to train an architecture of single and multiple-layer SNN on AudioMnist data set to 85% of test accuracy. We also mapped the model to analog hardware, reaching an accuracy of 72% for selected runs of the mapping protocol. In the paper, we share insights and considerations which affect the complicated procedure of mapping SNN models to analog custom hardware.