Mode hunting through active information

TL;DR

This paper introduces a novel mode detection method using active information, capable of identifying and locating modes in high-dimensional spaces without relying on PCA, and only detects modes when they truly exist.

Contribution

The paper presents an algorithm that detects and locates modes based on active information, reducing dimensionality without PCA and avoiding false positives.

Findings

Effective mode detection in high-dimensional spaces

Reduces dimensionality without PCA

Avoids false detection of non-existent modes

Abstract

We propose a new method to find modes based on active information. We develop an algorithm that, when applied to the whole space, will say whether there are any modes present \textit{and} where they are; this algorithm will reduce the dimensionality without resorting to Principal Components; and more importantly, population-wise, will not detect modes when they are not present.