Efficiency attacks on spiking neural networks
Sarada Krithivasan, Sanchari Sen, Nitin Rathi, Kaushik Roy, Anand Raghunathan
Abstract
Spiking Neural Networks are a class of artificial neural networks that process information as discrete spikes. The time and energy consumed in SNN implementations is strongly dependent on the number of spikes processed. We explore this sensitivity from an adversarial perspective and propose SpikeAttack, a completely new class of attacks on SNNs. SpikeAttack impacts the efficiency of SNNs via imperceptible perturbations that increase the overall spiking activity of the network, leading to increased time and energy consumption. Across four SNN benchmarks, SpikeAttackresults in 1.7x-2.5X increase in spike activity, leading to increases of 1.6x-2.3x and 1.4x-2.2x in latency and energy consumption, respectively.
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Your agent calls
Lunesearch_papers
Free to start. No credit card required.
Terminal
Install the CLIlune papers get eefb7b54-41c3-4cfc-9301-4cdad6e7217aCited by top-tier papers4
- LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless LoopsJiyuan Fu, Kaixun Jiang, Lingyi Hong, Jinglun Li et al.ICLR 2026 · 12 citations
- SoK: Efficiency Robustness of Dynamic Deep Learning SystemsRavishka Rathnasuriya, Tingxi Li, Zexin Xu, Zihe Song et al.USENIX Security 2025
- NaturalSloth: Revisiting Denial-of-Service Attacks on Large Language ModelsYiming Chen, Zexin Li, Xianghu Yue, Robby T. Tan et al.ACL 2026
- Towards Effective and Sparse Adversarial Attack on Spiking Neural Networks via Breaking Invisible Surrogate GradientsLi Lun, Kunyu Feng, Qinglong Ni, Ling Liang et al.CVPR 2025
Related papers
- EOS: An Energy-Oriented Attack Framework for Spiking Neural NetworksNing Yang, Fangxin Liu, Zongwu Wang, Haomin Li et al.DAC 2024 · 1 citation
- Rate Gradient Approximation Attack Threats Deep Spiking Neural NetworksTong Bu, Jianhao Ding, Zecheng Hao, Zhaofei YuCVPR 2023
- HIRE-SNN: Harnessing the Inherent Robustness of Energy-Efficient Deep Spiking Neural Networks by Training with Crafted Input NoiseSouvik Kundu, Massoud Pedram, Peter A. BeerelICCV 2021 · 114 citations
- Timestep-Compressed Attack on Spiking Neural Networks Through Timestep-Level BackpropagationDonghwa Kang, Doohyun Kim, Sang-Ki Ko, Jinkyu Lee et al.AAAI 2026
- On the Role of Temporal Granularity in the Robustness of Spiking Neural NetworksMengting Xu, Shi Gu, Peng Lin, De Ma et al.CVPR 2026
