Regularized Sparse Optimal Discriminant Clustering

Mayu Hiraishi, Kensuke Tanioka, Hiroshi Yadohisa · 2025

Read Regularized Sparse Optimal Discriminant Clustering by Mayu Hiraishi, Kensuke Tanioka, Hiroshi Yadohisa (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-10147.txt.

Description

Mayu Hiraishi, Kensuke Tanioka, and Hiroshi Yadohisa propose a clustering method in a twenty twenty-five paper. Clustering groups data points so that alike points sit together. Their base method uses a scoring matrix from discriminant analysis, kept sparse so the groups stay readable.

The new version adds a penalty drawn from convex clustering. The penalty pulls points in one cluster closer and pushes apart points from distinct clusters, which sharpens the picture when plotted. The scoring matrix must stay orthogonal while it learns this shape, a hard pair of demands. The team builds an update rule with a majorizing function and tests it on simulations and real data.

Catalog record

CREATORMayu Hiraishi, Kensuke Tanioka, Hiroshi Yadohisa
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE82 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-10147

Browse the full library · About the Tower of Babel