arXiv cs.AI· Lucas Senos Coutinho, Matheus Vargas, Pedro Teixeira, Heudson Tosta Mirandola, Ori Rottenstreich, Priscila Machado Vieira Lima, Leandro Santiago, Daniel Sadoc Menasche·· 4 小时前AI 评分31
二值化如何成为无权神经网络(WNN)的攻击面:以 WiSARD 为例的分析
Binarization as an Attack Surface in Weightless Neural Networks
AI 导读
一项发表于 ACSAC 2026 Trustworthy Machine Learning Workshop(ARTMAN 2026)的研究分析了 WiSARD 及其 Bloom filter 变体 Bloom WiSARD 与 BTHOWeN 中二值化带来的对抗攻击风险。
正文 · 原文
Abstract:Weightless Neural Networks (WNNs) are lightweight classifiers that binarize their inputs and learn by storing binary patterns in lookup memories, which makes them attractive for resource-constrained tasks such as intrusion detection. Because a small perturbation that crosses no binarization threshold leaves the binary input unchanged, WNNs may appear naturally robust to adversarial examples; a perturbation that does cross a threshold, however, can still change the prediction. We study this tension in WiSARD, the canonical RAM-based WNN, and in two variants that replace its memories with Bloom filters, Bloom WiSARD and BTHOWeN. A continuous perturbation reaches these classifiers only by crossing an encoding threshold, changing at least one binary tuple, and redirecting a memory query; we formalize this mechanism with a threshold-crossing bound and a response-margin certificate for standard Boolean WiSARD. Controlled MNIST experiments compare linear and data-driven thermometer encodings under model-agnostic sensitivity probes, a score-based Simulated Annealing attack, and surrogate-transfer attacks. Changing only the encoding while keeping each model otherwise fixed shows that encoding granularity can trade a small clean-accuracy loss for a large decrease in attack success. On the Bot-IoT, CICIDS, and UNSW-NB15 intrusion-detection datasets, adversarial perturbations lead WNN detectors to classify many attacks as normal traffic and to raise false alarms, even when their aggregate accuracy remains high.
| Comments: | This paper appears at the Trustworthy Machine Learning Workshop (ARTMAN 2026) co-located with the IEEE Annual Computer Security Applications Conference (ACSAC 2026) |
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.17646 [cs.CR] |
| (or arXiv:2611.17646v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17646 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Daniel Menasche [view email]
[v1]
Fri, 13 Nov 2026 01:34:07 UTC (14,938 KB)
来源:arXiv cs.AI · arxiv.org