In this paper we focus on label-flip attacks against Support Vector Machines (SVM), which are a state-of-the-art, widely used classifier. Previous work, ...
It is empirically show that the proposed defence strategy, referred to as Infinity-norm SVM, can significantly improve classifier security under malicious ...
Finally, we empirically show that the proposed defence strategy, referred to as Infinity-norm SVM, can significantly improve classifier security under malicious ...
This paper focuses on a strategy recently proposed in the literature to improve the robustness of linear classifiers to adversarial data manipulation.
In this work, we evaluate the security of Support Vector Machines (SVMs) to well-crafted, adversarial label noise attacks.
Missing: Infinity- Norm
The underlying rationale is to increase the number of support vectors and balance more equally their contribution to the decision function, to decrease the ...
Sep 3, 2019 · This work formulates a multistage game between an SVM and adversary. The SVM aims to maximize classification accuracy on a test dataset.
Infinity-norm support vector machines against adversarial label contamination. A Demontis, B Biggio, G Fumera, G Giacinto, F Roli. CEUR Workshop Proceedings ...
Infinity-norm support vector machines against adversarial label contamination. A Demontis, B Biggio, G Fumera, G Giacinto, F Roli. CEUR Workshop Proceedings ...
Infinity-norm Support Vector Machines against Adversarial Label Contamination. In A. Armando, R. Baldoni, and R. Focardi, editors, First Italian Conference on ...