Deep Spiking Neural Networks Based on STDP Rules for Image Recognition

Spiking neural networks (SNN) can transmit and process information through a sequence of spikes. Compared with the first two generations of artificial neural networks, spiking neural networks are closer to biological neurons and have high biological interpretability. In addition, SNN have low power consumption and good applications in neurochips. SNN learning algorithms include unsupervised algorithms and supervised algorithms. The STDP rule, as an archetype of unsupervised algorithms, is widely used and has been derived from several versions. This review concentrates on the application of STDP algorithm in the areas of image recognition.