Effects of Introducing Synaptic Scaling on Spiking Neural Network Learning

TL;DR

This paper explores how incorporating synaptic scaling into spiking neural networks enhances learning performance, demonstrating improved classification accuracy on MNIST datasets through specific normalization techniques.

Contribution

It introduces the integration of synaptic scaling with neural plasticity in WTA networks and evaluates its impact on classification accuracy using standard datasets.

Findings

L2-norm-based synaptic scaling improves accuracy

Achieved 88.84% on MNIST after one epoch

Achieved 68.01% on Fashion-MNIST after one epoch

Abstract

Spiking neural networks (SNNs) employing unsupervised learning methods inspired by neural plasticity are expected to be a new framework for artificial intelligence. In this study, we investigated the effect of multiple types of neural plasticity, such as spike-time-dependent plasticity (STDP) and synaptic scaling, on the learning in a winner-take-all (WTA) network composed of spiking neurons. We implemented a WTA network with multiple types of neural plasticity using Python. The MNIST and the Fashion-MNIST datasets were used for training and testing. We varied the number of neurons, the time constant of STDP, and the normalization method used in synaptic scaling to compare classification accuracy. The results demonstrated that synaptic scaling based on the L2 norm was the most effective in improving classification performance. By implementing L2-norm-based synaptic scaling and setting the number of neurons in both excitatory and inhibitory layers to 400, the network achieved classification accuracies of 88.84 % on the MNIST dataset and 68.01 % on the Fashion-MNIST dataset after one epoch of training.