How Hard Is Continuous Clustering? Lower Bounds from the Existential Theory of the Reals

2026-04-24

AI Summary

This study investigates the computational complexity of clustering problems defined over continuous probability densities specified by polynomials, focusing on four fundamental tasks: separating high-density points, detecting density valleys, counting connected components, and identifying holes. Leveraging tools from real algebraic geometry and the existential theory of the reals (ETR), the work provides the first rigorous complexity classification of exact continuous clustering within the real polynomial hierarchy. The main contributions establish that the first two tasks are ETR-complete, while the latter two are at least ETR-hard—their membership in ETR remains an open question. Furthermore, the paper demonstrates that these clustering criteria cannot be NP-complete unless an unexpected collapse of complexity classes occurs.

Abstract

This paper studies the computational difficulty of clustering problems that are defined directly on a continuous probability density. Rather than working with finite samples, we assume the density is given as a polynomial and ask whether it contains certain cluster structures. Four natural questions are examined. First, do there exist several points with high density that are far apart from each other. Second, do two high density points have a midpoint with low density, creating a valley between them. Third, does the region where the density is above a threshold have at least a given number of separate connected pieces. Fourth, does that same region contain a hole, meaning a loop that cannot be shrunk to a point. We prove that the first two problems, separated points and valley detection, are exactly as hard as the existential theory of the reals, a complexity class that contains NP and is believed to be strictly larger. In contrast, the topological problems of counting connected pieces and detecting holes are at least as hard as the existential theory of the reals, but their exact complexity remains open. Placing them inside that class would need a major advance in real algebraic geometry. These results give the first rigorous classification of exact continuous clustering inside the real polynomial hierarchy. They also show that even basic clustering criteria are not NP complete unless unexpected collapses occur.

Problem

• continuous clustering
• computational complexity
• existential theory of the reals
• topological structure
• density-based clustering

Innovation

• continuous clustering
• existential theory of the reals
• computational complexity
• real algebraic geometry
• topological clustering