Adversarial Robustness in Unsupervised Machine Learning: A Systematic Review

TL;DR

This systematic review examines the current state of adversarial robustness in unsupervised machine learning, highlighting prevalent attack types, defenses, and proposing a new model to guide future research in enhancing model security.

Contribution

The paper provides a comprehensive review of existing research and introduces a novel model outlining attack properties in unsupervised learning to aid future defense strategies.

Findings

Most research focuses on privacy attacks with effective defenses

Many attacks lack effective, general defensive measures

Proposes a new model for attack properties in unsupervised learning

Abstract

As the adoption of machine learning models increases, ensuring robust models against adversarial attacks is increasingly important. With unsupervised machine learning gaining more attention, ensuring it is robust against attacks is vital. This paper conducts a systematic literature review on the robustness of unsupervised learning, collecting 86 papers. Our results show that most research focuses on privacy attacks, which have effective defenses; however, many attacks lack effective and general defensive measures. Based on the results, we formulate a model on the properties of an attack on unsupervised learning, contributing to future research by providing a model to use.