Given a sample of points, all we get to see is who are the k nearest neighbors of every point. In other words, we get the adjacency matrix of the directed, ...

If we “zoom out very far”, the data will even look like a single point and thus may be consid- ered as zero-dimensional. While existing methods assume to know ...

This work provides two estimators for estimating the intrinsic dimension of datasets, a naive one and a more elaborate one that can be proved to be ...

TL;DR: This work provides two estimators for estimating the intrinsic dimension of datasets, a naive one and a more elaborate one that can be proved to be ...

Abstract. We address dimensionality estimation and nonlin- ear manifold inference starting from point inputs in high dimensional spaces using tensor voting.

In this paper, data dimensionality estimation methods are reviewed. The estima- tion of the dimensionality of a data set is a classical problem of pattern ...

The improved algorithm does not require a uniform truncation distance, that is, different sample points correspond to different truncation distances. According ...

Apr 21, 2024 · The estimates computed by non-parametric methods for outlier detection usually rely on the distances from the test point to its nearest ...

Dimensionality reduction methods are preprocessing techniques used for coping with high dimensionality. They have the aim of projecting the original data ...

Sep 22, 2017 · Here we propose a new ID estimator using only the distance of the first and the second nearest neighbor of each point in the sample.