Kernel t-distributed stochastic neighbor embedding

TL;DR

This paper introduces a kernelized t-SNE algorithm that maps high-dimensional data to low-dimensional space using kernel methods, enhancing clustering and data relationship visualization in non-Euclidean metrics.

Contribution

It proposes a novel kernelized extension of t-SNE capable of utilizing kernels in high-dimensional and low-dimensional spaces for improved data visualization.

Findings

Kernel t-SNE produces clearer clusters in datasets.

Enhanced visualization of data relationships.

Potential improvements in classification tasks.

Abstract

This paper presents a kernelized version of the t-SNE algorithm, capable of mapping high-dimensional data to a low-dimensional space while preserving the pairwise distances between the data points in a non-Euclidean metric. This can be achieved using a kernel trick only in the high dimensional space or in both spaces, leading to an end-to-end kernelized version. The proposed kernelized version of the t-SNE algorithm can offer new views on the relationships between data points, which can improve performance and accuracy in particular applications, such as classification problems involving kernel methods. The differences between t-SNE and its kernelized version are illustrated for several datasets, showing a neater clustering of points belonging to different classes.