Structured Prediction with Partial Labelling through the Infimum Loss - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Structured Prediction with Partial Labelling through the Infimum Loss

Prediction structurée à partir de données partiellement annotées via la "perte minimale"

Vivien Cabannes
  • Fonction : Auteur
  • PersonId : 1089863
Francis Bach
  • Fonction : Auteur
  • PersonId : 863086

Résumé

Annotating datasets is one of the main costs in nowadays supervised learning. The goal of weak supervision is to enable models to learn using only forms of labelling which are cheaper to collect, as partial labelling. This is a type of incomplete annotation where, for each datapoint, supervision is cast as a set of labels containing the real one. The problem of supervised learning with partial labelling has been studied for specific instances such as classification, multi-label, ranking or segmentation, but a general framework is still missing. This paper provides a unified framework based on structured prediction and on the concept of infimum loss to deal with partial labelling over a wide family of learning problems and loss functions. The framework leads naturally to explicit algorithms that can be easily implemented and for which proved statistical consistency and learning rates. Experiments confirm the superiority of the proposed approach over commonly used baselines.
Fichier principal
Vignette du fichier
main.pdf (546.54 Ko) Télécharger le fichier
tex_files.zip (1.41 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03127110 , version 1 (01-02-2021)

Licence

Paternité

Identifiants

  • HAL Id : hal-03127110 , version 1

Citer

Vivien Cabannes, Alessandro Rudi, Francis Bach. Structured Prediction with Partial Labelling through the Infimum Loss. ICML 2020 - 37th International Conference on Machine Learning, Jul 2020, Online, United States. pp.1230-1239. ⟨hal-03127110⟩
249 Consultations
69 Téléchargements

Partager

Gmail Facebook X LinkedIn More