English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The backpropagation algorithm has promoted the rapid development of deeplearning, but it relies on a large amount of labeled data and still has a largegap with how humans learn. The human brain can quickly learn various conceptualknowledge in a self-organized and unsupervised manner, accomplished throughcoordinating various learning rules and structures in the human brain.Spike-timing-dependent plasticity (STDP) is a general learning rule in thebrain, but spiking neural networks (SNNs) trained with STDP alone isinefficient and perform poorly. In this paper, taking inspiration fromshort-term synaptic plasticity, we design an adaptive synaptic filter andintroduce the adaptive spiking threshold as the neuron plasticity to enrich therepresentation ability of SNNs. We also introduce an adaptive lateralinhibitory connection to adjust the spikes balance dynamically to help thenetwork learn richer features. To speed up and stabilize the training ofunsupervised spiking neural networks, we design a samples temporal batch STDP(STB-STDP), which updates weights based on multiple samples and moments. Byintegrating the above three adaptive mechanisms and STB-STDP, our model greatlyaccelerates the training of unsupervised spiking neural networks and improvesthe performance of unsupervised SNNs on complex tasks. Our model achieves thecurrent state-of-the-art performance of unsupervised STDP-based SNNs in theMNIST and FashionMNIST datasets. Further, we tested on the more complex CIFAR10dataset, and the results fully illustrate the superiority of our algorithm. Ourmodel is also the first work to apply unsupervised STDP-based SNNs to CIFAR10.At the same time, in the small-sample learning scenario, it will far exceed thesupervised ANN using the same structure.

An Unsupervised STDP-based Spiking Neural Network Inspired By Biologically Plausible Learning Rules and Connections