# Efficiently Training Time-to-First-Spike Spiking Neural Networks from Scratch

Source: [Efficiently Training Time-to-First-Spike Spiking Neural Networks from Scratch](<https://www.sota2.com/research/paper/68fa775c1e209131cea72b6d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-28T05:00:05.261Z

Authors: Kaiwei Che, Zhengyu Ma, Yifan Huang, Peng Xue, Li Yuan, Timoth\\'ee Masquelier, Wei Fang, Yonghong Tian

Published: 2024-10-31T04:14:47.000Z

## Abstract

Spiking Neural Networks (SNNs), with their event-driven and biologically inspired mechanisms, are well-suited for energy-efficient neuromorphic hardware. Neural coding, which is critical to SNNs, determines how information is represented via spikes. While Time-to-First-Spike (TTFS) coding uses a single spike per neuron to offer extreme sparsity and energy efficiency, it often suffers from unstable training and low accuracy due to its sparse firing. To address these challenges, we propose a training framework that incorporates parameter initialization, training normalization, a temporal output decoder, and a re-evaluation of the pooling layer. The proposed parameter initialization and training normalization mitigate signal diminishing and gradient vanishing, which helps stabilize training. Our output decoder aggregates temporal spikes to encourage earlier firing, thereby reducing latency. The re-evaluation of the pooling layer demonstrates that max-pooling violates single-spike constraints, which should be avoided, whereas average-pooling preserves them. Experiments show that our framework stabilizes and accelerates training, reduces latency, and achieves state-of-the-art accuracy for step-by-step TTFS SNNs on MNIST (99.48%), Fashion-MNIST (92.90%), CIFAR10 (90.56%), CIFAR100 (70.27%) and DVS Gesture (95.83%).

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Classification on CIFAR10 (test) (Efficiency Metrics: Time-steps, SOPs, Energy)](<https://www.sota2.com/research/sota/image-classification-on-cifar10-test-efficiency-metrics-time-steps-sops-energy.md>): Image Classification; CIFAR10 (test); Time-steps: 2.93.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2410.23619>)

[PDF](<https://arxiv.org/pdf/2410.23619.pdf>)
