面向人工神经网络（ANNs）与脉冲神经网络（SNNs）的表达能力归一化能耗对比

中文摘要 AI 辅助

脉冲神经网络（SNNs）常被视为人工神经网络（ANNs）的高能效替代方案，但其优势高度依赖网络架构与数据特性。我们开发了一个分析框架，针对时间序列数据，在匹配表达能力的条件下对比全连接ReLU型ANNs与积分放电型SNNs的理论能效。通过将推理能耗模型与表征表达能力的理论界值关联，我们推导了表达能力归一化能效比，以及网络宽度、脉冲稀疏度和ANN深度缩放的明确阈值。本分析明确了事件驱动计算抵消SNNs时间开销的场景，为设计高能效时间网络提供了感知容量的原则，结果表明仅在特定场景下ANNs的表达能力归一化能效超过SNNs。

英文摘要

Spiking neural networks (SNNs) are often regarded as energy-efficient alternatives to artificial neural networks (ANNs), yet their advantage depends critically on both network architecture and data properties. We develop an analytical framework to compare fully-connected ReLU ANNs and integrate-and-fire SNNs for time-series data with respect to their theoretical energy efficiency at matched expressive capacity. By relating an inference-energy model to theoretical bounds on representational expressivity, we derive an expressivity-normalized efficiency ratio and explicit thresholds in network width, spike sparsity, and ANN depth scaling. Our analysis characterizes the regimes in which event-driven computation offsets the temporal overhead of SNNs, providing capacity-aware principles for designing energy-efficient temporal networks. It shows that ANNs exceed SNNs in expressivity-normalized efficiency only in specific regimes.

发表机构

• LMU Munich（慕尼黑大学）
• University of Tromsø（特罗姆瑟大学）
• DLR-German Aerospace Center（德国航空航天中心）
• Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）

机构由 AI 辅助整理，请以论文原文为准。