This paper describes some algorithms for the following problem : given a nonlinear function defined as a black box, find which variables the function actually depends on (the set of active variables) and for each of these variables test if the relation is linear or nonlinear. Alternatively, verify if a given set of active variables and/or a given linearity pattern is correct. The algorithms are based on group testing and binary search and they are...
Voir la suite
INFORMATION
-
1979/08/31
-
Document de travail (série numérotée)
-
DRC14
-
1
-
1
-
2020/06/13
-
Disclosed
-
Detecting or verifying sparsity structure in large scale nonlinear programming
-
blood sample
TÉLÉCHARGER
RAPPORT COMPLET
Version officielle du document (peut inclure des signatures etc…)
-
Total Downloads** : 73
*La version texte est une version à reconnaissance optique de caractères non-corrigée. Cette version est fournie uniquement pour accommoder les utilisateurs disposant de connections lentes.