Yann Guermeur

I am currently at the head of the   ABC   research team of the   LORIA-UMR 7503   in Nancy
Fédération Charles Hermite

"Directeur de recherche" CNRS
"Notice de titres et travaux"




PhD Students

Hafida Bouziane-Chouarfia (2008-2014)
Rémi Bonidal (2009-2013)
Mounia Hendel (2010-)
Edouard Klein (2011-2013)




Events

Special session "Supervised Prediction with Neural Networks and SVMs" of ASMDA 2007 Conference

Pascal Theoretical Challenge. Type I and type II errors for multiple simultaneous hypothesis testing




Research Projects and Working Groups

PASCAL 2

Projet GENOTO3D de l'ACI "Masses de Données"

GdT "Apprentissage et séquences" de l'AS CNRS ASAB

Project Last update
GENOTO3D 11-24-06
ASAB 04-08-04




Software

Software of the M-SVM2 (Frank-Wolfe algorithm)
Software of the M-SVM2 (Rosen's algorithm)
MSVMpack
MSVMpred

Application Last update
M-SVM2 (Frank-Wolfe algorithm) 04-04-10
M-SVM2 (Rosen's algorithm) 11-16-10
MSVMpack 03-29-11



Teaching

Multi-Class Support Vector Machines (pdf). Summer School NN2008, Porto.
Multi-Class Support Vector Machines (ps). Summer School NN2008, Porto.
Protein Secondary Structure Prediction with Multi-Class Support Vector Machines (pdf). Summer School NN2008, Porto.
Protein Secondary Structure Prediction with Multi-Class Support Vector Machines (ps). Summer School NN2008, Porto.



Some papers

Full publication list

SVM Multiclasses, Théorie et Applications. Y. Guermeur (2007). HDR, Université Nancy 1.

Combinaison de classifieurs statistiques, application à la prédiction de la structure secondaire des protéines. Y. Guermeur (1997). PhD thesis, Université Paris 6.

Estimation et contrôle des performances en généralisation des réseaux de neurones. Y. Guermeur and O. Teytaud (2006). In Y. Bennani editor, Apprentissage Connexionniste, Chap. 10, 279-342, Hermès.

A kernel for protein secondary structure prediction. Y. Guermeur, A. Lifchitz and R. Vert (2004). In B. Schölkopf, K. Tsuda and J.-P. Vert, editors, Kernel Methods in Computational Biology, Chap. 9, 193-206, the MIT Press.

Théorie de l'apprentissage de Vapnik et SVM, Support Vector Machines. Y. Guermeur and H. Paugam-Moisy (1999). In M. Sebban and G. Venturini, editors, Apprentissage Automatique, 109-138, Hermès.

Comments on: Support vector machines maximizing geometric margins for multi-class classification. Y. Guermeur (2014). TOP, Vol. 22, N. 3, 844-851.

Model selection for the l2-SVM by following the regularization path. R. Bonidal, S. Tindel, and Y. Guermeur (2014). Transactions on Computational Collective Intelligence (TCCI), Vol. XIII (LNCS 8342), 83-112.

Combining multi-class SVMs with linear ensemble methods that estimate the class posterior probabilities. Y. Guermeur (2013). Communications in Statistics - Theory and Methods, Vol. 42, N. 16, 3011-3030.

A generic model of multi-class support vector machine. Y. Guermeur (2012). International Journal of Intelligent Information and Database Systems (IJIIDS), Vol. 6, N. 6, 555-577.

Hybrid feature selection and SVM-based classification for mouse skin precancerous stages diagnosis from bimodal spectroscopy. F. Abdat, M. Amouroux, Y. Guermeur, and W. Blondel (2012). Optics Express, Vol. 20, N. 1, 228-244.

MSVMpack: a multi-class support vector machine package. F. Lauer and Y. Guermeur (2011). Journal of Machine Learning Research (JMLR), Vol. 12, 2293-2296.

A quadratic loss multi-class SVM for which a radius-margin bound applies. Y. Guermeur and E. Monfrini (2011). INFORMATICA, Vol. 22, N. 1, 73-96.

Sample complexity of classifiers taking values in RQ, application to multi-class SVMs. Y. Guermeur (2010). Communications in Statistics - Theory and Methods, Vol. 39, N. 3, 543-557.

HECTAR: A method to predict subcellular targeting in heterokonts. B. Gschloessl, Y. Guermeur and J.M. Cock (2008). BMC Bioinformatics, Vol. 9, 393.

VC theory of large margin multi-category classifiers. Y. Guermeur (2007). JMLR, Vol. 8, 2551-2594.

Prediction of amphipathic in-plane membrane anchors in monotopic proteins using a SVM classifier. N. Sapay, Y. Guermeur and G. Deléage (2006). BMC Bioinformatics, Vol. 7, 255.

A comparative study of multi-class support vector machines in the unifying framework of large margin classifiers. Y. Guermeur, A. Elisseeff and D. Zelus (2005). Applied Stochastic Models in Business and Industry (ASMBI), Vol. 21, N. 2, 199-214.

Combining protein secondary structure prediction models with ensemble methods of optimal complexity. Y. Guermeur, G. Pollastri, A. Elisseeff, D. Zelus, H. Paugam-Moisy and P. Baldi (2004). Neurocomputing, Vol. 56, 305-327.

Combining discriminant models with new multi-class SVMs. Y. Guermeur (2002). Pattern Analysis and Applications (PAA), Vol. 5, N. 2, 168-179.

Improved performance in protein secondary structure prediction by inhomogeneous score combination. Y. Guermeur, C. Geourjon, P. Gallinari and G. Deléage (1999). Bioinformatics, Vol. 15, N. 5, 413-421.

Twelve numerical, symbolic and hybrid supervised classification methods. O. Gascuel et al. (1998). International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI), Vol. 12, N. 5, 517-571.

Etude comparée des performances de SVM multi-classes en prédiction de la structure secondaire des protéines. Y. Guermeur (2009). Revue des Nouvelles Technologies de l'Information (RNTI), Vol. A-3, 21-48.

Théorie de l'apprentissage de Vapnik et SVM, Support Vector Machines. Y. Guermeur and H. Paugam-Moisy (1999). Revue Electronique sur l'Apprentissage par les Données (READ), Vol. 3, N. 1, 17-38.

Cascading Discriminant and Generative Models for Protein Secondary Structure Prediction. F. Thomarat, F. Lauer, and Y. Guermeur (2012). PRIB'12, Tokyo, 166-177.

Estimating the class posterior probabilities in biological sequence segmentation. R. Bonidal, F. Thomarat, and Y. Guermeur (2012). SMTDA'12, Chania.

Estimating the class posterior probabilities in protein secondary structure prediction. Y. Guermeur and F. Thomarat (2011). PRIB'11, Delft, 260-271.

DCT-SVM based multi-classification of mouse skin precancerous stages from autofluorescence and diffuse reflectance spectra. F. Abdat, M. Amouroux, Y. Guermeur and W. Blondel (2011). ECBO'11, Munich.

Ensemble Methods of Appropriate Capacity for Multi-Class Support Vector Machines. Y. Guermeur (2010). SMTDA'10, Chania, 311-318.

Radius-Margin Bound on the Leave-One-Out Error of the LLW-M-SVM. Y. Guermeur and E. Monfrini (2009). ASMDA'09, Vilnius, 517-521.

Scale-sensitive Psi-dimensions: the Capacity Measures for Classifiers Taking Values in RQ. Y. Guermeur (2007). ASMDA'07, Chania.

Model selection for multi-class SVMs. Y. Guermeur, M. Maumy and F. Sur (2005). ASMDA'05, Brest, 507-517.

Bound on the risk for M-SVMs. Y. Guermeur, A. Elisseeff and D. Zelus (2002). Statistical Learning, Theory and Applications, Paris, 48-52.

A new multi-class SVM based on a uniform convergence result. Y. Guermeur, A. Elisseeff and H. Paugam-Moisy (2000). IJCNN'00, Come, Vol. IV, 183-188.

Generalization performance of multi-class discriminant models. H. Paugam-Moisy, A. Elisseeff and Y. Guermeur (2000). IJCNN'00, Come, Vol. IV, 177-182.

Estimating the sample complexity of a multi-class discriminant model. Y. Guermeur, A. Elisseeff and H. Paugam-Moisy (1999). ICANN'99, Edimbourg, 310-315.

Multivariate Linear Regression on Classifier Outputs: a Capacity Study. Y. Guermeur, H. Paugam-Moisy and P. Gallinari (1998). ICANN'98, Skövde, 693-698.

Optimal Linear Regression on Classifier Outputs. Y. Guermeur, F. d'Alché-Buc and P. Gallinari (1997). ICANN'97, Lausanne, 481-486.

Combining Statistical Models for Protein Secondary Structure Prediction. Y. Guermeur and P. Gallinari (1996). ICANN'96, Bochum, 599-604.

Approach based on artificial neural networks for protein secondary structure prediction. H. Bouziane-Chouarfia, B. Messabih, Y. Guermeur and A. Chouarfia (2009). NTICRI'09, Oran.

Borne "rayon-marge" sur l'erreur "leave-one-out" des SVM multi-classes. Y. Darcy, E. Monfrini and Y. Guermeur (2006). JdS'06, Clamart.

Prediction of in-plane amphipathic membrane segments based on an SVM method. N. Sapay, Y. Guermeur and G. Deléage (2005). JOBIM'05, Lyon, 299-311.

Traitement Statistique des Résultats de SELEX. D. Eveillard and Y. Guermeur (2002). JOBIM'02, Saint-Malo, 277-283.

Combining protein secondary structure prediction models with ensemble methods of optimal complexity. Y. Guermeur and D. Zelus (2001). JOBIM'01, Toulouse, 97-104.

Risque garanti pour les modèles de discrimination multi-classes. A. Elisseeff, H. Paugam-Moisy and Y. Guermeur (1999). SFC'99 , Nancy, 111-118.

Combinaison de classifieurs estimant les probabilités a posteriori des classes. Y. Guermeur (1998). SFC'98 , Montpellier, 121-124.

Combinaison Linéaire Optimale de Classifieurs. Y. Guermeur, F. d'Alché-Buc and P. Gallinari (1997). JdS'97, Carcassonne, 425-428.

Radius-Margin Bound on the Leave-One-Out Error of the LLW-M-SVM. Y. Guermeur and E. Monfrini (2009). Research Report LORIA.

A Quadratic Loss Multi-Class SVM. E. Monfrini and Y. Guermeur (2008). Research Report LORIA, hal-00276700.

Radius-margin Bound on the Leave-one-out Error of Multi-class SVMs. Y. Darcy and Y. Guermeur (2005). Research Report INRIA, RR-5780.

Large Margin Multi-category Discriminant Models and Scale-sensitive Psi-dimensions. Y. Guermeur (2004). Research Report INRIA, RR-5314 (revised in 2006).

Recherche des gènes d'ARN non codant. E. Gothié, Y. Guermeur, S. Muller, C. Branlant and A. Bockmayr (2003). Research Report INRIA, RR-5057.

A Simple Unifying Theory of Multi-Class Support Vector Machines. Y. Guermeur (2002). Research Report INRIA, RR-4669.

Bounding the Capacity Measure of Multi-Class Discriminant Models. Y. Guermeur, A. Elisseeff and D. Zelus (2002). Technical Report NeuroCOLT2, 2002-123.

Combining discriminant models with new multi-class SVMs. Y. Guermeur (2000). Technical Report NeuroCOLT2, 2000-086.

Margin error and generalization capabilities of multi-class discriminant systems. A. Elisseeff, Y. Guermeur and H. Paugam-Moisy (1999). Technical Report NeuroCOLT2, 1999-051.

Margin error and generalization capabilities of multi-class discriminant systems. Revised manuscript (draft 06-01).

Linear Ensemble Methods for Multiclass Discrimination. Y. Guermeur and H. Paugam-Moisy (1998). Research Report 1998-52, LIP, ENS Lyon.

New cascade architecture for protein secondary structure prediction. Y. Guermeur and F. Thomarat (2010). Abstract of a poster presented at JOBIM'10, Montpellier.

Statistical Processing of SELEX Results. D. Eveillard and Y. Guermeur (2002). Abstract of a poster presented at ISMB'2002, Edmonton.

Combining Protein Secondary Structure Prediction Methods with a new Multi-Category SVM. Y. Guermeur and D. Zelus (2000). Abstract of a poster presented at ISMB'2000, San Diego.

An Ensemble Method for Protein Secondary Structure Prediction. Y. Guermeur, F. d'Alché-Buc and P. Gallinari (1997). Abstract of an oral presentation at MABS'97, Rouen.



Yann Guermeur

Last modified 11-21-2013

e-mail: Yann.Guermeur@loria.fr