Fast-SNN: Fast Spiking Neural Network by Converting Quantized ANN

TL;DR

Fast-SNN introduces a novel conversion method from quantized ANNs to SNNs, significantly reducing inference latency and achieving state-of-the-art performance in vision tasks by minimizing quantization and sequential errors.

Contribution

The paper presents a new ANN-to-SNN conversion approach that minimizes quantization and sequential errors, enabling high-performance, low-latency SNNs for vision applications.

Findings

Achieves state-of-the-art results on vision tasks

Reduces inference latency significantly

Introduces a signed IF neuron model and layer-wise fine-tuning

Abstract

Spiking neural networks (SNNs) have shown advantages in computation and energy efficiency over traditional artificial neural networks (ANNs) thanks to their event-driven representations. SNNs also replace weight multiplications in ANNs with additions, which are more energy-efficient and less computationally intensive. However, it remains a challenge to train deep SNNs due to the discrete spike function. A popular approach to circumvent this challenge is ANN-to-SNN conversion. However, due to the quantization error and accumulating error, it often requires lots of time steps (high inference latency) to achieve high performance, which negates SNN's advantages. To this end, this paper proposes Fast-SNN that achieves high performance with low latency. We demonstrate the equivalent mapping between temporal quantization in SNNs and spatial quantization in ANNs, based on which the minimization of the quantization error is transferred to quantized ANN training. With the minimization of the quantization error, we show that the sequential error is the primary cause of the accumulating error, which is addressed by introducing a signed IF neuron model and a layer-wise fine-tuning mechanism. Our method achieves state-of-the-art performance and low latency on various computer vision tasks, including image classification, object detection, and semantic segmentation. Codes are available at: https://github.com/yangfan-hu/Fast-SNN.