Plus de mathématiques pour gagner plus

Apprentissage de distance pour l’annotation d’images par plus proches voisins

Image annotation is an important open problem in computer vision. For this task we propose TagProp, a weighted nearest neighbor model, discriminatively trained to exploit labeled training images for predicting tags of test images. Neighbor weights are based on neighbor rank or distance. TagProp can learn the metric that defines neighbors by maximizing the log-likelihood of the tag predictions i...

Experience plus explanation

The more clearly the clinician comprehends the processes of clinical perception and of linking the experience with explanation, the more thoughtfully clinical students can be guided to construct an accurate 'clinical memory' of cases, and to construct a sound 'working knowledge' structure from their personal 'examined experience'. The structuring of the steps of clinical learning is intended as...

ÉCOLE DOCTORALE DE MATHÉMATIQUES ET D ’ INFORMATIQUE THÈSE Pour obtenir le grade de DOCTEUR DE L ' UNIVERSITÉ BORDEAUX 1

........................................................................................................................................iii Résumé ..........................................................................................................................................iv Table of

Thèse de doctorat Discipline: Mathématiques

This thesis is concerned with the development and application of categorical tools in the study of the cluster algebras of S. Fomin and A. Zelevinsky. C. Amiot’s generalized cluster category is a triangulated category which has been used, in the case where it is Hom-finite, to categorify a certain class of cluster algebras, using cluster characters in the sense of Y. Palu. In this thesis, we ge...

Algorithmique de la réduction de réseaux et application à la recherche de pires cas pour l'arrondi defonctions mathématiques