Show simple item record

dc.creatorHernández Alvarado, Alberto José
dc.creatorSolís Chacón, Maikol
dc.creatorZúñiga Rojas, Ronald Alberto
dc.date.accessioned2020-03-17T17:06:22Z
dc.date.available2020-03-17T17:06:22Z
dc.date.issued2019
dc.identifier.citationhttp://jmlr.org/papers/v20/
dc.identifier.issn1533-7928
dc.identifier.issn1532-4435
dc.identifier.urihttps://hdl.handle.net/10669/80729
dc.description.abstractThe curse of dimensionality is a commonly encountered problem in statistics and data analysis. Variable sensitivity analysis methods are a well studied and established set of tools designed to overcome these sorts of problems. However, as this work shows, these methods fail to capture relevant features and patterns hidden within the geometry of the enveloping manifold projected onto a variable. Here we propose an index that captures, reflects and correlates the relevance of distinct variables within a model by focusing on the geometry of their projections. We construct the 2-simplices of a Vietoris-Rips complex and then estimate the area of those objects from a data-set cloud. The analysis was made with an original R-package called TopSA, short for Topological Sensitivity Analysis. The TopSA R-package is available at the site https://github.com/maikol-solis/TopSA.es_ES
dc.language.isoen_USes_ES
dc.sourceJournal of Machine Learning Research, vol.20, pp.1-21es_ES
dc.subjectGoodness of fites_ES
dc.subjectR2es_ES
dc.subjectVietoris-Rip complexes_ES
dc.subjectManifoldses_ES
dc.subjectArea estimationes_ES
dc.titleGeometric goodness of fit measure to detect patterns in data point cloudses_ES
dc.typepreprint
dc.description.procedenceUCR::Vicerrectoría de Investigación::Unidades de Investigación::Ciencias Básicas::Centro de investigaciones Matemáticas y Metamatemáticas (CIMM)es_ES
dc.description.procedenceUCR::Vicerrectoría de Investigación::Unidades de Investigación::Ciencias Básicas::Centro de Investigaciones en Matemáticas Puras y Aplicadas (CIMPA)es_ES


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record