Supervised And Aggregate-Label Learning Algorithm of Spiking Neural Network

• Published in: International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) (Online) pp. 33 - 36
• Main Authors: CHANG, HAOYANG, LI, JIANPING, FENG, MINGCHAO
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2019
• Subjects: Aggregate-label learning; Biological neural networks; Biological system modeling; Encoding; Neuroscience; Spiking neural networks; Spiking neurons; Supervised learning; Synaptic weight
• ISBN: 9781728142418; 1728142415
• ISSN: 2576-8964
• DOI: 10.1109/ICCWAMTIP47768.2019.9067538

Spiking Neural Networks (SNNs) is closer to biological neural working mechanisms. Compared with the traditional neural network using rate coding, SNNs are able to process and abstract features from the temporal dynamics encoded in spike signals, thus prompting SNNs more biologically plausible and easier to implement on hardware. This paper describes the existing supervised and aggregate-label classic learning algorithms of SNNs, analyzes the merits and demerits of these algorithms, discusses the development direction of SNNs, and provides a basis for the research of efficient learning algorithms.