Adversarial Examples for $k$-Nearest Neighbor Classifiers Based on Higher-Order Voronoi Diagrams

Abstract

Adversarial examples are a widely studied phenomenon in machine learning models. While most of the attention has been focused on neural networks, other practical models also suffer from this issue. In this work, we propose an algorithm for evaluating the adversarial robustness of $k$-nearest neighbor classification, i.e., finding a minimum-norm adversarial example. Diverging from previous proposals, we take a geometric approach by performing a search that expands outwards from a given input point. On a high level, the search radius expands to the nearby Voronoi cells until we find a cell that classifies differently from the input point. To scale the algorithm to a large $k$, we introduce approximation steps that find perturbations with smaller norm, compared to the baselines, in a variety of datasets. Furthermore, we analyze the structural properties of a dataset where our approach outperforms the competition.

arXiv commentAppears at NeurIPS 2021. Code is available at https://github.com/wagner-group/geoadex

References in corpus (8)

• Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning
• Evaluating Robustness of Neural Networks with Mixed Integer Programming
• Towards Fast Computation of Certified Robustness for ReLU Networks
• Analyzing the Robustness of Nearest Neighbors to Adversarial Examples
• Provable Certificates for Adversarial Examples: Fitting a Ball in the Union of Polytopes
• Defending Against Adversarial Examples with K-Nearest Neighbor
• Robustness for Non-Parametric Classification: A Generic Attack and Defense
• Evaluating the Robustness of Nearest Neighbor Classifiers: A Primal-Dual Perspective

Cited by in corpus (1)

• Degree distributions in AB random geometric graphs