Recognizing patterns in images using a small spiking neural network
We study a spiking neural network consisting of a small number of LIF neurons, capable of recognizing horizontal and vertical patterns in a 3*3 pixel image. Neural learning is carried out using the STDP algorithm. The network is trained and tested on a sample of ${6 0 0 0}$ images. It is shown that the investigated neural network has an accuracy of ${9 9 \%}$ on the training set and ${9 7 \%}$ on the test set.
Authors
- Dmitry M. Ezhov (ORCID: https://orcid.org/0000-0002-3311-5088)
- В. И. Пономаренко (ORCID: https://orcid.org/0000-0002-1579-6465)
- М. Д. Прохоров (ORCID: https://orcid.org/0000-0003-4069-9410)
- A. V. Kurbako (ORCID: https://orcid.org/0000-0002-3479-4609)
Institutions
- Saratov State University (RU)
- Institute of Physics (RU)
Publication Details
- Published
- 2024-09-19
- DOI
- https://doi.org/10.1109/dcna63495.2024.10718500
- Primary Topic
- Neural Networks and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Russian Science Foundation