VMF-SNE: Embedding for Spherical Data

TL;DR

This paper introduces vMF-SNE, a new embedding algorithm tailored for spherical data, which outperforms traditional t-SNE in visualizing high-dimensional spherical datasets.

Contribution

The paper proposes a novel vMF-SNE algorithm that models local proximity with von Mises-Fisher distributions, improving embedding quality for spherical data.

Findings

vMF-SNE produces better embeddings than t-SNE for spherical data

The iterative process is efficient for high-dimensional spherical data

Simulation results validate the superiority of vMF-SNE in relevant scenarios

Abstract

T-SNE is a well-known approach to embedding high-dimensional data and has been widely used in data visualization. The basic assumption of t-SNE is that the data are non-constrained in the Euclidean space and the local proximity can be modelled by Gaussian distributions. This assumption does not hold for a wide range of data types in practical applications, for instance spherical data for which the local proximity is better modelled by the von Mises-Fisher (vMF) distribution instead of the Gaussian. This paper presents a vMF-SNE embedding algorithm to embed spherical data. An iterative process is derived to produce an efficient embedding. The results on a simulation data set demonstrated that vMF-SNE produces better embeddings than t-SNE for spherical data.