Document Type

Article

Publication Date

2022

Publisher

Journal of Machine Learning Research (JMLR)

Source Publication

Journal of Machine Learning Research (JMLR)

Source ISSN

1532-4435

Abstract

In Topological Data Analysis, a common way of quantifying the shape of data is to use a persistence diagram (PD). PDs are multisets of points in R2 computed using tools of algebraic topology. However, this multi-set structure limits the utility of PDs in applications. Therefore, in recent years efforts have been directed towards extracting informative and efficient summaries from PDs to broaden the scope of their use for machine learning tasks. We propose a computationally efficient framework to convert a PD into a vector in Rn, called a vectorized persistence block (VPB). We show that our representation possesses many of the desired properties of vector-based summaries such as stability with respect to input noise, low computational cost and flexibility. Through simulation studies, we demonstrate the effectiveness of VPBs in terms of performance and computational cost for various learning tasks, namely clustering, classification and change point detection.

Comments

Published version. Journal of Machine Learning Research (JMLR), Vol. 23 (2022). Publisher link. © 2022 Journal of Machine Learning Research (JMLR). Used with permission.

CC-BY 4.0, see https://creativecommons.org/licenses/by/4.0/.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Available for download on Thursday, January 02, 2025

Included in

Mathematics Commons

Share

COinS