Meyers, Charles and Löfstedt, Tommy and Elmroth, Erik (2024) Massively Parallel Evasion Attacks and the Pitfalls of Adversarial Retraining. EAI Endorsed Transactions on Internet of Things.
![[thumbnail of 70965.pdf]](http://eprints.eai.eu/style/images/fileicons/application_pdf.png "70965.pdf")
PDF
70965.pdf
Download (1MB)
Abstract
Even with widespread adoption of automated anomaly detection in safety-critical areas, both classical and advanced machine learning models are susceptible to first-order evasion attacks that fool models at run-time (e.g. an automated firewall or an anti-virus application). Kernelized support vector
| Item Type: | Article |
|---|---|
| Date Deposited: | 04 Mar 2026 18:15 |
| Last Modified: | 16 Apr 2026 22:06 |
| URI: | http://eprints.eai.eu/id/eprint/51796 |
Actions (login required)
- View Item