Rethinking the Energy Efficiency of SNNs and ANNs: A Perspective from Neural Network Design
ACM Transactions on Architecture and Code Optimization · 30.09.2026 · 23 (3) · 1-26
DOI: 10.1145/3822176 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 24.07.2026. Verfasst von Miao Yu und 2 weiteren Autor:innen. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: A*STAR under the RIE2025 Energy-aware Accelerated Computing (EAC) program, RIE2025 Innovation & Enterprise (I&E) Industry Alignment Fund - Industry Collaboration Projects, Ministry of Education, Singapore, Tier 3.
Abstract (Originalsprache)
Spiking Neural Networks (SNNs) have shown great potential for energy-efficient computing. However, current SNN designs primarily focus on synaptic operations as the main energy metric, often overlooking the impact of memory access when compared with Artificial Neural Networks (ANNs). To help researchers design energy-efficient SNNs, we propose a novel analytical methodology aimed at comparing the energy consumption and latency of SNNs and ANNs based on validated low-level details from accelerator prototypes. This work explores how neural network design influences energy consumption and latency, a previously underexplored area, to uncover counterintuitive results that highlight opportunities for optimizing the energy and latency of SNNs. Based on the results obtained from our methodology, we conclude that (a) SNN architectures have the potential to achieve better energy efficiency with comparable accuracy when handling tasks with Transformer-based systems and Multi-Layer Perceptrons. However, (b) when handling complex neural networks and datasets, the energy efficiency and latency of SNNs are affected by the network topology, timestep, and firing rate. Furthermore, (c) if we take into account pruning techniques, SNNs remain less efficient than traditional ANN systems for CNN architectures. Experiments show that pruned SNNs still consume 1.61× more energy than pruned ANNs. Consequently, most of the existing SNN algorithms still lag behind ANNs in terms of energy efficiency and latency due to their large number of timesteps, and high firing rate in modern networks. We open-source the methodology and examples at https://github.com/nus-comparch/SNN-vs-ANN to support reproducibility and allow the broader community to use this work.
Förderung
- A*STAR under the RIE2025 Energy-aware Accelerated Computing (EAC) program
- RIE2025 Innovation & Enterprise (I&E) Industry Alignment Fund - Industry Collaboration Projects
- Ministry of Education, Singapore, Tier 3
- National University of Singapore under FD-fAbrICS: Joint Lab for FD-SOI Always-on Intelligent & Connected Systems
- NUS Artificial Intelligence Institute
Zitieren
APA
Yu, M., Xiang, T., & Carlson, T. E. (2026). Rethinking the Energy Efficiency of SNNs and ANNs: A Perspective from Neural Network Design. ACM Transactions on Architecture and Code Optimization, 23(3), 1-26. https://doi.org/10.1145/3822176
Vancouver
Yu M, Xiang T, Carlson TE. Rethinking the Energy Efficiency of SNNs and ANNs: A Perspective from Neural Network Design. ACM Transactions on Architecture and Code Optimization. 2026;23(3):1-26. doi:10.1145/3822176
BibTeX
@article{Yu2026,
author = {Yu, Miao and Xiang, Tingting and Carlson, Trevor E.},
title = {Rethinking the Energy Efficiency of SNNs and ANNs: A Perspective from Neural Network Design},
journal = {ACM Transactions on Architecture and Code Optimization},
year = {2026},
volume = {23},
number = {3},
pages = {1--26},
doi = {10.1145/3822176}
}
EUVIMED