GRIBONVAL, Rémi; BLANCHARD, Gilles; KERIVEN, Nicolas; TRAONMILIN, Yann

doi:10.4171/msl/20

Metadatos

Licencia de uso del documento

GRIBONVAL, Rémi
Parcimonie et Nouveaux Algorithmes pour le Signal et la Modélisation Audio [PANAMA]
Dynamic Networks : Temporal and Structural Capture Approach [DANTE]

BLANCHARD, Gilles
Institut für Mathematik [Potsdam]
Understanding the Shape of Data [DATASHAPE]
Laboratoire de Mathématiques d'Orsay [LMO]

KERIVEN, Nicolas
Parcimonie et Nouveaux Algorithmes pour le Signal et la Modélisation Audio [PANAMA]
GIPSA Pôle Géométrie, Apprentissage, Information et Algorithmes [GIPSA-GAIA]

Idioma

Article de revue

Este ítem está publicado en

Mathematical Statistics and Learning. 2021-08-21, vol. 3, n° 2, p. 113–164

EMS Publishing House

Resumen en inglés

We describe a general framework --compressive statistical learning-- for resource-efficient large-scale learning: the training collection is compressed in one pass into a low-dimensional sketch (a vector of random empirical generalized moments) that captures the information relevant to the considered learning task. A near-minimizer of the risk is computed from the sketch through the solution of a nonlinear least squares problem. We investigate sufficient sketch sizes to control the generalization error of this procedure. The framework is illustrated on compressive PCA, compressive clustering, and compressive Gaussian mixture Modeling with fixed known variance. The latter two are further developed in a companion paper.< Leer menos

Palabras clave en inglés

Dimension reduction

statistical learning

Kernel mean embedding

sketching

random features

excess risk control

random moments

URI

https://oskar-bordeaux.fr/handle/20.500.12278/191535

DOI

http://dx.doi.org/10.4171/msl/20

Proyecto europeo

PLEASE: Projections, Learning, and Sparsity for Efficient data-processing

Proyecto ANR

Algorithmes, Approximations, Parcimonie et Plongements pour l'IA - ANR-19-CHIA-0009
Approches statistiquement et computationnellement efficicaces pour l'intelligence artificielle - ANR-19-CHIA-0021

Orígen

Importado de Hal

Centros de investigación

Institut de Mathématiques de Bordeaux (IMB) - UMR 5251