A fast homotopy algorithm for a large class of weighted classification problems - Archive ouverte HAL Accéder directement au contenu
Poster De Conférence Année : 2013

A fast homotopy algorithm for a large class of weighted classification problems

Résumé

We propose a penalized method to solve the one-way ANOVA problem by collapsing the coefficients of K conditions. We introduce a large class of weights for which our homotopy algorithm is in O(K log(K)). These weights induce a balanced tree structure and simplify the interpretation of the results. As an example we consider phenotypic data: given a trait, we reconstruct a balanced tree structure and assess its agreement with the known phylogeny. Our proposal is easily extended to more than one dimension for clustering problems.
Fichier principal
Vignette du fichier
MLCB13_submission4_1.pdf (231.85 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-01601843 , version 1 (03-06-2020)

Identifiants

  • HAL Id : hal-01601843 , version 1
  • PRODINRA : 396493

Citer

Pierre Gutierrez, Guillem Rigaill, Julien Chiquet. A fast homotopy algorithm for a large class of weighted classification problems. MLCB NIPS 2013, Oct 2013, South Lake Thao, United States. 2013. ⟨hal-01601843⟩
59 Consultations
11 Téléchargements

Partager

Gmail Facebook X LinkedIn More