Deep Learning in Spiking Phasor Neural Networks

TL;DR

This paper introduces Spiking Phasor Neural Networks (SPNNs), a complex-valued neural network model that uses spike timing to encode phases, achieving high performance on CIFAR-10 and bridging the gap between SNNs and traditional DNNs.

Contribution

The paper presents a novel SPNN model that employs complex-valued representations and spike timing coding, advancing the capabilities of SNNs in image classification tasks.

Findings

SPNNs outperform other timing coded SNNs on CIFAR-10

Performance of SPNNs approaches that of real-valued DNNs

Demonstrates robust computation using spike timing code

Abstract

Spiking Neural Networks (SNNs) have attracted the attention of the deep learning community for use in low-latency, low-power neuromorphic hardware, as well as models for understanding neuroscience. In this paper, we introduce Spiking Phasor Neural Networks (SPNNs). SPNNs are based on complex-valued Deep Neural Networks (DNNs), representing phases by spike times. Our model computes robustly employing a spike timing code and gradients can be formed using the complex domain. We train SPNNs on CIFAR-10, and demonstrate that the performance exceeds that of other timing coded SNNs, approaching results with comparable real-valued DNNs.