On Tree-based Methods for Similarity Learning - Télécom Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

On Tree-based Methods for Similarity Learning

Résumé

In many situations, the choice of an adequate similarity measure or metric on the feature space dramatically determines the performance of machine learning methods. Building automatically such measures is the specific purpose of metric/similarity learning. In [21], similarity learning is formulated as a pairwise bipartite ranking problem: ideally, the larger the probability that two observations in the feature space belong to the same class (or share the same label), the higher the similarity measure between them. From this perspective, the ROC curve is an appropriate performance criterion and it is the goal of this article to extend recur-sive tree-based ROC optimization techniques in order to propose efficient similarity learning algorithms. The validity of such iterative partitioning procedures in the pairwise setting is established by means of results pertaining to the theory of U-processes and from a practical angle, it is discussed at length how to implement them by means of splitting rules specifically tailored to the similarity learning task. Beyond these theoret-ical/methodological contributions, numerical experiments are displayed and provide strong empirical evidence of the performance of the algorith-mic approaches we propose.
Fichier principal
Vignette du fichier
1906.09243.pdf (684.42 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02461801 , version 1 (30-01-2020)

Identifiants

  • HAL Id : hal-02461801 , version 1

Citer

Stéphan Clémençon, Robin Vogel. On Tree-based Methods for Similarity Learning. The Fifth Conference on Machine Learning, Optimization, and Data Science (LOD 2019), Sep 2019, Sienne, Italy, Italy. pp.676-688. ⟨hal-02461801⟩
36 Consultations
140 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More