Slice inverse regression with score functions - Archive ouverte HAL Access content directly
Journal Articles Electronic Journal of Statistics Year : 2018

Slice inverse regression with score functions


We consider non-linear regression problems where we assume that the response depends non-linearly on a linear projection of the covariates. We propose score function extensions to sliced inverse regression problems, both for the first-order and second-order score functions. We show that they provably improve estimation in the population case over the non-sliced versions and we study finite sample estimators and their consistency given the exact score functions. We also propose to learn the score function as well, in two steps, i.e., first learning the score function and then learning the effective dimension reduction space, or directly, by solving a convex optimization problem regularized by the nuclear norm. We illustrate our results on a series of experiments.
Fichier principal
Vignette du fichier
Sliced Inverse regression with score functions (1).pdf (667.64 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01388498 , version 1 (27-10-2016)
hal-01388498 , version 2 (21-11-2018)



Dmitry Babichev, Francis Bach. Slice inverse regression with score functions. Electronic Journal of Statistics , 2018, Volume 12, Number 1 (2018), pp.1507-1543. ⟨10.1214/18-EJS1428⟩. ⟨hal-01388498v2⟩
232 View
1097 Download



Gmail Facebook Twitter LinkedIn More