Learning and Real-time Classification of Hand-written Digits With Spiking Neural Networks

TL;DR

This paper presents a novel spiking neural network that efficiently classifies handwritten digits in real-time with high accuracy, leveraging GPU implementation and a new supervised learning algorithm, achieving state-of-the-art performance with fewer parameters.

Contribution

The paper introduces a new SNN architecture with a NormAD-based learning algorithm, optimized for real-time digit classification on GPU with significantly fewer parameters.

Findings

Achieved 98.06% accuracy on MNIST test set.

Real-time inference with over 97% accuracy within 100ms.

Reduced model complexity by nearly 7x compared to existing SNNs.

Abstract

We describe a novel spiking neural network (SNN) for automated, real-time handwritten digit classification and its implementation on a GP-GPU platform. Information processing within the network, from feature extraction to classification is implemented by mimicking the basic aspects of neuronal spike initiation and propagation in the brain. The feature extraction layer of the SNN uses fixed synaptic weight maps to extract the key features of the image and the classifier layer uses the recently developed NormAD approximate gradient descent based supervised learning algorithm for spiking neural networks to adjust the synaptic weights. On the standard MNIST database images of handwritten digits, our network achieves an accuracy of 99.80% on the training set and 98.06% on the test set, with nearly 7x fewer parameters compared to the state-of-the-art spiking networks. We further use this network in a GPU based user-interface system demonstrating real-time SNN simulation to infer digits written by different users. On a test set of 500 such images, this real-time platform achieves an accuracy exceeding 97% while making a prediction within an SNN emulation time of less than 100ms.