Geometric Algorithms for k-NN Poisoning

06/21/2023

∙

We propose a label poisoning attack on geometric data sets against k-nearest neighbor classification. We provide an algorithm that can compute an ε n-additive approximation of the optimal poisoning in n· 2^2^O(d+k/ε) time for a given data set X ∈ℝ^d, where |X| = n. Our algorithm achieves its objectives through the application of multi-scale random partitions.

READ FULL TEXT

Geometric Algorithms for k-NN Poisoning

Online nearest neighbor classification

Euclidean TSP, Motorcycle Graphs, and Other New Applications of Nearest-Neighbor Chains

All nearest neighbor calculation based on Delaunay graphs

Nearest Neighbor Sampling of Point Sets using Random Rays

Multi-scale Template Matching with Scalable Diversity Similarity in an Unconstrained Environment

Neighborhood Growth Determines Geometric Priors for Relational Representation Learning

Certifiable Robustness for Nearest Neighbor Classifiers

Geometric Algorithms for k-NN Poisoning

Related Research

Online nearest neighbor classification

Euclidean TSP, Motorcycle Graphs, and Other New Applications of Nearest-Neighbor Chains

All nearest neighbor calculation based on Delaunay graphs

Nearest Neighbor Sampling of Point Sets using Random Rays

Multi-scale Template Matching with Scalable Diversity Similarity in an Unconstrained Environment

Neighborhood Growth Determines Geometric Priors for Relational Representation Learning

Certifiable Robustness for Nearest Neighbor Classifiers