在树莓派上高效运行脉冲神经网络，实现92%准确率
Spiffy: Efficient Implementation of CoLaNET for Raspberry Pi
- 用Rust重写CoLaNET架构，无需专用硬件
- 训练每步仅0.9毫秒，推理每步0.45毫秒
- 适合边缘计算与低功耗设备开发者
本文提出一种轻量级软件方案，在不依赖专用类脑硬件或框架的前提下运行脉冲神经网络（SNN）。我们采用Rust语言实现特定的SNN架构CoLaNET，并针对通用计算平台进行优化。以树莓派为案例，展示名为Spiffy的实现，在MNIST数据集上达到92%的准确率，训练每步延迟仅0.9毫秒，推理每步延迟0.45毫秒。代码已开源。
原文摘要 · Abstract (English)
This paper presents a lightweight software-based approach for running spiking neural networks (SNNs) without relying on specialized neuromorphic hardware or frameworks. Instead, we implement a specific SNN architecture (CoLaNET) in Rust and optimize it for common computing platforms. As a case study, we demonstrate our implementation, called Spiffy, on a Raspberry Pi using the MNIST dataset. Spiffy achieves 92% accuracy with low latency - just 0.9 ms per training step and 0.45 ms per inference step. The code is open-source.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。