Paper published in a book (Scientific congresses and symposiums)
Nets versus trees for feature ranking and gene network inference
Vecoven, Nicolas; Begon, Jean-Michel; Sutera, Antonio et al.
2020In Proceeding of the 23rd International Conference on Discovery Science (DS 2020)
Peer reviewed
 

Files


Full Text
Vecoven2020_Chapter_NetsVersusTreesForFeatureRanki.pdf
Publisher postprint (513 kB)
Request a copy

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Feature ranking; Deep learning; Neural networks; Gene regulatory network inference; Random forests
Abstract :
[en] We investigate several global variable importance measures derived from artificial neural networks (ANN) to address the challenging problem of feature ranking in high-dimensional unstructured problems. While several ANN (local) importance measures have been validated in the context of computer vision or natural language processing tasks, it is not clear how these methods perform on unstructured problems where many variables are expected to be irrelevant. We empirically compare these ANN measures with one standard and state-of-the-art Random forests (RF) importance measure on several artificial and real datasets. These experiments show that ANN measures can achieve performance similar to the RF measure, sometimes outperforming it. On some problems however, the feature rankings returned by ANN are not as good as the ones returned by RF, despite significantly better predictive performance. Importantly, reaching the best performance with the ANN-based methods often comes at the cost of introducing a so-called selection layer at the beginning of the network. Using this specific neural architecture has proven to be critical both in terms of feature ranking and predictive performance on datasets with many irrelevant variables. Finally, we evaluate these methods on the problem of gene network inference, where they yield decent performance, without however outperforming RF.
Disciplines :
Computer science
Author, co-author :
Vecoven, Nicolas ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
Begon, Jean-Michel ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Algorith. des syst. en interaction avec le monde physique
Sutera, Antonio ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Méthodes stochastiques
Geurts, Pierre ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Algorith. des syst. en interaction avec le monde physique
Huynh-Thu, Vân Anh ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes microélectroniques intégrés
Language :
English
Title :
Nets versus trees for feature ranking and gene network inference
Publication date :
2020
Event name :
23rd International Conference on Discovery Science (DS 2020)
Event date :
from 19-10-2020 to 21-10-2020
Audience :
International
Main work title :
Proceeding of the 23rd International Conference on Discovery Science (DS 2020)
Publisher :
Springer
Peer reviewed :
Peer reviewed
Available on ORBi :
since 28 October 2020

Statistics


Number of views
103 (18 by ULiège)
Number of downloads
10 (10 by ULiège)

Scopus citations®
 
1
Scopus citations®
without self-citations
1
OpenCitations
 
2

Bibliography


Similar publications



Contact ORBi