Adversarial Classification: Necessary conditions and geometric flows

11/21/2020
by   Nicolas Garcia Trillos, et al.
0

We study a version of adversarial classification where an adversary is empowered to corrupt data inputs up to some distance ε, using tools from variational analysis. In particular, we describe necessary conditions associated with the optimal classifier subject to such an adversary. Using the necessary conditions, we derive a geometric evolution equation which can be used to track the change in classification boundaries as ε varies. This evolution equation may be described as an uncoupled system of differential equations in one dimension, or as a mean curvature type equation in higher dimension. In one dimension we rigorously prove that one can use the initial value problem starting from ε=0, which is simply the Bayes classifier, in order to solve for the global minimizer of the adversarial problem. Numerical examples illustrating these ideas are also presented.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/04/2019

Perspectives on the Formation of Peakons in the Stochastic Camassa-Holm Equation

The stochastic Camassa-Holm equation was derived in Holm and Tyranowski ...
research
04/04/2021

Tukey Depths and Hamilton-Jacobi Differential Equations

The widespread application of modern machine learning has increased the ...
research
05/05/2022

Geometric Methods for Adjoint Systems

Adjoint systems are widely used to inform control, optimization, and des...
research
05/10/2021

Scalar auxiliary variable approach for conservative/dissipative partial differential equations with unbounded energy

In this paper, we present a novel investigation of the so-called SAV app...
research
05/21/2018

The limit shape of convex hull peeling

We prove that the convex peeling of a random point set in dimension d ap...
research
05/18/2021

Curvature-dimension conditions for symmetric quantum Markov semigroups

Following up on the recent work on lower Ricci curvature bounds for quan...
research
03/16/2022

Characterization and modeling of the influence of initial microstructure on recrystallization of zircaloy-4 during hot forming

The present article proposes a detailed study of recrystallization of zi...

Please sign up or login with your details

Forgot password? Click here to reset