FAKERI TABRIZI Ali
Supervision : Patrick GALLINARI
Co-supervision : AMINI Massih-Reza
Semi-supervised Multi-view Learning: An Application to Image Annotation and Multi-lingual Document Classification
In this thesis, we introduce two multiview learning approaches. In a first approach, we describe a self-training multiview strategy which trains different voting classifiers on different views. The margin distributions over the unlabeled training data, obtained with each view-specific classifier are then used to estimate an upper-bound on their transductive Bayes error. Minimizing this upper-bound provides an automatic margin-threshold which is used to assign pseudo-labels to unlabeled examples. Final class labels are then assigned to these examples, by taking a vote on the pool of the previous pseudo-labels. New view-specific classifiers are then trained using the original labeled and the pseudo-labeled training data. We consider applications to image-text and to multilingual document classification.
In second approach, we propose a multiview semi-supervised bipartite ranking model which allows us to leverage the information contained in unlabeled sets of images to improve the prediction performance, using multiple descriptions, or views of images. For each topic class, our approach first learns as many view-specific rankers as there are available views using the labeled data only. These rankers are then improved iteratively by adding pseudo-labeled pairs of examples on which all view-specific rankers agree over the ranking of examples within these pairs.
We report on experiments carried out on the NUS-WIDE dataset, which show that the multiview ranking process improves predictive performance when a small number of labeled examples is available specially for unbalanced topic classes. We show also that our approach achieves significant improvements over a state-of-the art semi-supervised multiview classification model. We present experimental results on the NUS-WIDE collection and on Reuters RCV1-RCV2 which show that despite its simplicity, our approach is competitive with other state-of-the-art techniques.
Defence : 09/30/2013
Jury members :
Mr. Glotin, Hérvé. Univ. Sud Toulon Var [Rapporteur]
Mr. Quenot, Georges. Office B-109 Campus Scientifique [Rapporteur]
Mr. Artières, Thierry. LIP6, Université Pierre et Marie Curie
Mr. Gallinari, Patrick. LIP6, Université Pierre et Marie Curie
Mr. Amini, Massih-reza. LIG/AMA
2008-2015 Publications
-
2015
- A. Fakeri‑Tabrizi, M.‑R. Amini, C. Goutte, N. Usunier : “Multiview self-learning”, Neurocomputing, vol. 155, pp. 117–127, (Elsevier) (2015)
-
2013
- A. Fakeri Tabrizi : “Apprentissage Semi-supervisĂ© Multi-vues: Une Application pour L’annotation d’Image et la Classification de Documents Multilingues”, thesis, phd defence 09/30/2013, supervision Gallinari, Patrick, co-supervision : Amini, Massih-Reza (2013)
- A. Fakeri‑Tabrizi, M.‑R. Amini, P. Gallinari : “Multiview Semi-Supervised Ranking for Automatic Image Annotation”, ACM International Conference on Multimedia, Barcelone, Spain, pp. 513-516, (ACM) (2013)
-
2011
- H. Sahbi, X. Li, S. Tollari, H. Glotin, Ph. Mulhem, A. Fakeri‑Tabrizi, Zh. Zhao, P. Gallinari : “Apprentissage automatique pour l’annotation d’images”, chapitre de SĂ©mantique et multimodalitĂ© en analyse de l'information, TraitĂ© RTA, sĂ©rie Recherche d'information et web, pp. 367-381, (Hermès-Lavoisier), (ISBN: 2-7462-3139-5) (2011)
-
2010
- A. Fakeri‑Tabrizi, S. Tollari, N. Usunier, P. Gallinari : “UPMC/LIP6 at ImageCLEFannotation 2010”, Working Notes for CLEF 2010 Conference, vol. 1176, CEUR-WS, Padua, Italy, (CEUR) (2010)
- A. Fakeri‑Tabrizi, S. Tollari, N. Usunier, P. Gallinari : “Improving Image Annotation in Imbalanced Classification Problems with Ranking SVM”, Multilingual Information Access Evaluation II. Multimedia Experiments, vol. 6242, Lecture Notes in Computer Science, Corfu, Greece, pp. 291-294, (Springer) (2010)
- S. Tollari, M. Detyniecki, A. Fakeri‑Tabrizi, Ch. Marsala, M.‑R. Amini, P. Gallinari : “Exploitation du contenu visuel pour amĂ©liorer la recherche textuelle d’images en ligne”, Document numĂ©rique - Revue des sciences et technologies de l'information. SĂ©rie Document numĂ©rique, vol. 13 (1), pp. 187-209, (Hermès) (2010)
-
2009
- H. Glotin, A. Fakeri‑Tabrizi, Ph. Mulhem, M. Ferecatu, Zh. Zhao, S. Tollari, G. Quenot, H. Sahbi, E. Dumont, P. Gallinari : “Comparison of Various AVEIR Visual Concept Detectors with an Index of Carefulness”, Image Cross Language Evaluation Forum (ImageClef 2009), vol. 1175, CEUR Workshop Proceedings, Corfu, Greece, (ceur-ws.org) (2009)
- S. Tollari, M. Detyniecki, A. Fakeri‑Tabrizi, Ch. Marsala, M.‑R. Amini, P. Gallinari : “Using Visual Concepts and Fast Visual Diversity to Improve Image Retrieval”, 9th Workshop of the Cross-Language Evaluation Forum, CLEF 2008, Revised Selected Papers, vol. 5706, Lecture Notes in Computer Science, Aarhus, Denmark, pp. 577-584, (Springer Berlin / Heidelberg) (2009)
- A. Fakeri‑Tabrizi, S. Tollari, L. Denoyer, P. Gallinari : “UPMC/LIP6 at ImageCLEFannotation 2009: Large Scale Visual Concept Detection and Annotation”, CLEF working notes 2009, Corfu, Greece (2009)
- S. Tollari, M. Detyniecki, A. Fakeri‑Tabrizi, Ch. Marsala, M.‑R. Amini, P. Gallinari : “Utilisation de concepts visuels et de la diversitĂ© visuelle pour amĂ©liorer la recherche d’images”, Actes de ConfĂ©rence en Recherche d'Informations et Applications (CORIA'09), Toulon, France, pp. 83-98 (2009)
- S. Tollari, M. Detyniecki, Ch. Marsala, A. Fakeri‑Tabrizi, M.‑R. Amini, P. Gallinari : “Exploiting Visual Concepts to Improve Text-Based Image Retrieval”, European Conference on Information Retrieval (ECIR), vol. 5478, Lecture Notes in Computer Science, Toulouse, France, pp. 701-705, (Springer) (2009)
-
2008
- S. Tollari, M. Detyniecki, A. Fakeri‑Tabrizi, M.‑R. Amini, P. Gallinari : “UPMC/LIP6 at ImageCLEFphoto 2008: on the exploitation of visual concepts (VCDT)”, QUAERO/ImageCLEF Workshop on Multimedia Information Retrieval Evaluation, Aarhus, Denmark, pp. 1-10 (2008)
- S. Tollari, M. Detyniecki, M. Ferecatu, H. Glotin, Ph. Mulhem, M.‑R. Amini, A. Fakeri‑Tabrizi, P. Gallinari, H. Sahbi, Zh. Zhao : “Consortium AVEIR at ImageCLEFphoto 2008: on the fusion of runs”, Working Notes for the CLEF 2008 workshop, vol. 1174, CEUR-WS, Aarhus, Denmark, (CEUR) (2008)
- A. Fakeri‑Tabrizi, M.‑R. Amini, S. Tollari, P. Gallinari : “UPMC/LIP6 at ImageCLEF’s WikipediaMM: An Image-Annotation Model for an Image Search-Engine”, Working Notes for the CLEF 2008 workshop, vol. 1174, CEUR-WS, Aarhus, Denmark, (CEUR) (2008)