English

Kernel $k$-Medoids as General Vector Quantization

Machine Learning 2025-09-08 v2 Quantum Physics

Abstract

Vector Quantization (VQ) is a widely used technique in machine learning and data compression, valued for its simplicity and interpretability. Among hard VQ methods, kk-medoids clustering and Kernel Density Estimation (KDE) approaches represent two prominent yet seemingly unrelated paradigms -- one distance-based, the other rooted in probability density matching. In this paper, we investigate their connection through the lens of Quadratic Unconstrained Binary Optimization (QUBO). We compare a heuristic QUBO formulation for kk-medoids, which balances centrality and diversity, with a principled QUBO derived from minimizing Maximum Mean Discrepancy in KDE-based VQ. Surprisingly, we show that the KDE-QUBO is a special case of the kk-medoids-QUBO under mild assumptions on the kernel's feature map. This reveals a deeper structural relationship between these two approaches and provides new insight into the geometric interpretation of the weighting parameters used in QUBO formulations for VQ.

Keywords

Cite

@article{arxiv.2506.04786,
  title  = {Kernel $k$-Medoids as General Vector Quantization},
  author = {Thore Gerlach and Sascha Mücke and Christian Bauckhage},
  journal= {arXiv preprint arXiv:2506.04786},
  year   = {2025}
}
R2 v1 2026-07-01T03:00:58.033Z