A Simple Geometric Interpretation of SVM using Stochastic Adversaries

17 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

On Security and Sparsity of Linear Classifiers for Adversarial Settings (2016, background): “Connecting the work in [13,14,15] to adversarial machine learning aims to help understanding what the optimal reg-ularizer is against diﬀerent kinds of adversarial noise (attacks).”
On Robustness and Regularization of Structural Support Vector Machines (2014, background): “Livni et al. (2012) show that a minimax formulation of robustness in the presence of stochastic adversaries results in L2 (Frobenius for matrix weights) regularization, and for the multi-class case results in twoinfinity regularization of the model weights. Torkamani and Lowd (2013), show that for a”
Secure Kernel Machines against Evasion Attacks (2016, background): “We clarify here the connection between regularization and input data uncertainty highlighted by the recent ﬁndings in [17, 20, 23, 30, 34].”