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

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.
Complete list of metadatas

https://hal.univ-cotedazur.fr/hal-01322765
Contributor : Denis Pallez <>
Submitted on : Friday, May 27, 2016 - 4:17:16 PM
Last modification on : Monday, November 5, 2018 - 3:52:10 PM

Identifiers

Collections

Citation

R. Pighetti, Denis Pallez, Frédéric Precioso. Improving SVM Training Sample Selection Using Multi-Objective Evolutionary Algorithm and LSH. Computational Intelligence, 2015 IEEE Symposium Series on, 2015, Cape Town, South Africa. pp.1383-1390, ⟨10.1109/SSCI.2015.197⟩. ⟨hal-01322765⟩

Share

Metrics

Record views

221