# Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics

## TL;DR

This paper introduces a visual analytics framework that helps users understand and explore vulnerabilities of machine learning models to adversarial attacks, aiding in the defense against malicious manipulations.

## Contribution

The paper presents a novel visual analytics framework specifically designed for analyzing and explaining model vulnerabilities to adversarial attacks in machine learning.

## Key findings

- Framework effectively visualizes data poisoning attack vulnerabilities
- Case studies demonstrate insights into attack strategies and model weaknesses
- Supports analysis from multiple perspectives including data, features, and local structures

## Abstract

Machine learning models are currently being deployed in a variety of real-world applications where model predictions are used to make decisions about healthcare, bank loans, and numerous other critical tasks. As the deployment of artificial intelligence technologies becomes ubiquitous, it is unsurprising that adversaries have begun developing methods to manipulate machine learning models to their advantage. While the visual analytics community has developed methods for opening the black box of machine learning models, little work has focused on helping the user understand their model vulnerabilities in the context of adversarial attacks. In this paper, we present a visual analytics framework for explaining and exploring model vulnerabilities to adversarial attacks. Our framework employs a multi-faceted visualization scheme designed to support the analysis of data poisoning attacks from the perspective of models, data instances, features, and local structures. We demonstrate our framework through two case studies on binary classifiers and illustrate model vulnerabilities with respect to varying attack strategies.