Detecting Potential Local Adversarial Examples for Human-Interpretable Defense

TL;DR

This paper proposes a method to detect potential local adversarial examples in tabular data by identifying critical features influencing the classifier's decision, aiding human experts in fraud prevention.

Contribution

It introduces a novel approach to identify locally critical features for classifiers, enhancing interpretability and fraud detection in tabular data.

Findings

Initial proposition for detecting local adversarial examples

Provides critical features to human experts for decision control

Aims to improve fraud detection accuracy

Abstract

Machine learning models are increasingly used in the industry to make decisions such as credit insurance approval. Some people may be tempted to manipulate specific variables, such as the age or the salary, in order to get better chances of approval. In this ongoing work, we propose to discuss, with a first proposition, the issue of detecting a potential local adversarial example on classical tabular data by providing to a human expert the locally critical features for the classifier's decision, in order to control the provided information and avoid a fraud.