Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH

R. Pighetti; Denis Pallez; Frédéric Precioso

doi:10.1109/SSCI.2015.197

Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH

R. Pighetti Denis Pallez ¹ Frédéric Precioso ²

1 I3S - Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis

2 Laboratoire d'Informatique, Signaux, et Systèmes de Sophia-Antipolis (I3S) / Equipe KEIA

SPARKS - Scalable and Pervasive softwARe and Knowledge Systems

Abstract : In this paper, we propose a new framework hybridizing a Support Vector Machine (SVM), a Multi-Objective Genetic Algorithm (MOGA) and a Locality Sensitive Hashing (LSH). The goal is to tackle fine-grained classification challenges which means classifying many classes with high similarities between classes and poor similarities inside one class. SVM is used for its ability of learning multi-class problem from very few training data. MOGA is used for optimizing training samples used by SVM so as to improve its learning rate. As data define a discrete set of instances and not a continuous solution space, LSH is used for mapping "optimal solutions" obtained by MOGA onto the closest real instances contained in the dataset. We evaluate our method for content-based image classification on the standard image database Caltech256 (i.e. 30000 images distributed in 256 classes). Experiments shows that our method outperforms reference approaches.

Keywords : Training data Training Buildings support vector machine multiobjective genetic algorithm Context Evolutionary computation genetic algorithms image classification support vector machines LSH MOGA SVM training sample selection content-based image classification fine-grained classification locality sensitive hashing multiobjective evolutionary algorithm

Type de document :

Communication dans un congrès

Computational Intelligence, 2015 IEEE Symposium Series on, 2015, Cape Town, South Africa. pp.1383-1390, 2015, Computational Intelligence and Data Mining. 〈10.1109/SSCI.2015.197〉

Domaine :

Informatique [cs] / Recherche opérationnelle [cs.RO]

Informatique [cs] / Apprentissage [cs.LG]

Liste complète des métadonnées

https://hal.univ-cotedazur.fr/hal-01322765
Contributeur : Denis Pallez <>
Soumis le : vendredi 27 mai 2016 - 16:17:16
Dernière modification le : mercredi 31 janvier 2018 - 10:24:05

Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH

Identifiants

Collections

Citation

Exporter

Métriques

111

Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH

Identifiants

Collections

Citation

Exporter

Partager

Métriques

111