Filter-based feature selection has become crucial in many classification settings, especially object recognition, recently faced with feature learning strategies that originate thousands of cues. In this paper, we propose a feature selection method exploiting the convergence properties of power series of matrices, and introducing the concept of infinite feature selection (Inf-FS). Considering a selection of features as a path among feature distributions and letting these paths tend to an infinite number permits the investigation of the importance (relevance and redundancy) of a feature when injected into an arbitrary set of cues. Ranking the importance individuates candidate features, which turn out to be effective from a classification point of view, as proved by a thoroughly experimental section. The Inf-FS has been tested on thirteen diverse benchmarks, comparing against filters, embedded methods, and wrappers; in all the cases we achieve top performances, notably on the classification tasks of PASCAL VOC 2007-2012.

Roffo, G., Melzi, S., Cristiani, M. (2015). Infinite Feature Selection. In 15th IEEE International Conference on Computer Vision, ICCV 2015 (pp.4202-4210). Institute of Electrical and Electronics Engineers Inc. [10.1109/ICCV.2015.478].

Infinite Feature Selection

Melzi, S;
2015

Abstract

Filter-based feature selection has become crucial in many classification settings, especially object recognition, recently faced with feature learning strategies that originate thousands of cues. In this paper, we propose a feature selection method exploiting the convergence properties of power series of matrices, and introducing the concept of infinite feature selection (Inf-FS). Considering a selection of features as a path among feature distributions and letting these paths tend to an infinite number permits the investigation of the importance (relevance and redundancy) of a feature when injected into an arbitrary set of cues. Ranking the importance individuates candidate features, which turn out to be effective from a classification point of view, as proved by a thoroughly experimental section. The Inf-FS has been tested on thirteen diverse benchmarks, comparing against filters, embedded methods, and wrappers; in all the cases we achieve top performances, notably on the classification tasks of PASCAL VOC 2007-2012.
No
paper
Feature Selection; Variable Ranking; Feature Ranking; Centrality; Graph Theory;
English
15th IEEE International Conference on Computer Vision, ICCV 2015
978-1-4673-8391-2
Roffo, G., Melzi, S., Cristiani, M. (2015). Infinite Feature Selection. In 15th IEEE International Conference on Computer Vision, ICCV 2015 (pp.4202-4210). Institute of Electrical and Electronics Engineers Inc. [10.1109/ICCV.2015.478].
Roffo, G; Melzi, S; Cristiani, M
File in questo prodotto:
File Dimensione Formato  
07410835.pdf

Solo gestori archivio

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Dimensione 221.61 kB
Formato Adobe PDF
221.61 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/350586
Citazioni
  • Scopus 230
  • ???jsp.display-item.citation.isi??? 186
Social impact