Spatial Transformer K-Means

TL;DR

This paper introduces a novel K-means clustering method that maintains data interpretability and invariance to non-rigid transformations, improving performance and providing convergence guarantees.

Contribution

It proposes a similarity measure invariant to non-rigid transformations, enabling intrinsic data space clustering with enhanced interpretability and theoretical guarantees.

Findings

Achieves state-of-the-art clustering performance.

Provides convergence guarantees for the clustering algorithm.

Enables interpretable clustering directly in the input data space.

Abstract

K-means defines one of the most employed centroid-based clustering algorithms with performances tied to the data's embedding. Intricate data embeddings have been designed to push $K$-means performances at the cost of reduced theoretical guarantees and interpretability of the results. Instead, we propose preserving the intrinsic data space and augment K-means with a similarity measure invariant to non-rigid transformations. This enables (i) the reduction of intrinsic nuisances associated with the data, reducing the complexity of the clustering task and increasing performances and producing state-of-the-art results, (ii) clustering in the input space of the data, leading to a fully interpretable clustering algorithm, and (iii) the benefit of convergence guarantees.