Current browse context:
math.AT
Change to browse by:
References & Citations
Mathematics > Algebraic Topology
Title: Reconstruction of univariate functions from directional persistence diagrams
(Submitted on 3 Mar 2022 (v1), last revised 8 Feb 2023 (this version, v2))
Abstract: We describe a method for approximating a single-variable function $f$ using persistence diagrams of sublevel sets of $f$ from height functions in different directions. We provide algorithms for the piecewise linear case and for the smooth case. Three directions suffice to locate all local maxima and minima of a piecewise linear continuous function from its collection of directional persistence diagrams, while five directions are needed in the case of smooth functions with non-degenerate critical points.
Our approximation of functions by means of persistence diagrams is motivated by a study of importance attribution in machine learning, where one seeks to reduce the number of critical points of signal functions without a significant loss of information for a neural network classifier.
Submission history
From: Carles Casacuberta [view email][v1] Thu, 3 Mar 2022 18:19:54 GMT (677kb,D)
[v2] Wed, 8 Feb 2023 22:17:57 GMT (678kb,D)
Link back to: arXiv, form interface, contact.