Scalable Optimal Classifiers for Adversarial Settings under Uncertainty

06/28/2021
by   Patrick Loiseau, et al.
0

We consider the problem of finding optimal classifiers in an adversarial setting where the class-1 data is generated by an attacker whose objective is not known to the defender – an aspect that is key to realistic applications but has so far been overlooked in the literature. To model this situation, we propose a Bayesian game framework where the defender chooses a classifier with no a priori restriction on the set of possible classifiers. The key difficulty in the proposed framework is that the set of possible classifiers is exponential in the set of possible data, which is itself exponential in the number of features used for classification. To counter this, we first show that Bayesian Nash equilibria can be characterized completely via functional threshold classifiers with a small number of parameters. We then show that this low-dimensional characterization enables to develop a training method to compute provably approximately optimal classifiers in a scalable manner; and to develop a learning algorithm for the online setting with low regret (both independent of the dimension of the set of possible data). We illustrate our results through simulations.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/18/2020

Protecting Classifiers From Attacks. A Bayesian Approach

Classification problems in security settings are usually modeled as conf...
research
02/21/2023

Characterizing the Optimal 0-1 Loss for Multi-class Classification with a Test-time Attacker

Finding classifiers robust to adversarial examples is critical for their...
research
02/13/2021

Mixed Nash Equilibria in the Adversarial Examples Game

This paper tackles the problem of adversarial examples from a game theor...
research
07/17/2022

Achieve Optimal Adversarial Accuracy for Adversarial Deep Learning using Stackelberg Game

Adversarial deep learning is to train robust DNNs against adversarial at...
research
12/13/2012

Learning Sparse Low-Threshold Linear Classifiers

We consider the problem of learning a non-negative linear classifier wit...
research
07/12/2020

BaCOUn: Bayesian Classifers with Out-of-Distribution Uncertainty

Traditional training of deep classifiers yields overconfident models tha...
research
06/29/2023

Towards Optimal Randomized Strategies in Adversarial Example Game

The vulnerability of deep neural network models to adversarial example a...

Please sign up or login with your details

Forgot password? Click here to reset