My work in the area of machine learning focuses on novel learning settings such as semi-supervised learning and structured prediction as well as (kernel-based) data fusion. A lot of my work rests on convex optimization techniques and spectral methods.
Publications:
| Title | Authors | Status | Year |
|---|---|---|---|
| An end-to-end machine learning system for harmonic analysis of music | Ni Yizhao, Matt McVicar, Raul Santos-Rodriguez, Tijl De Bie | journal paper | 2012 |
| Learning to Translate: A Statistical and Computational Analysis | Marco Turchi, Tijl De Bie, Cyril Goutte, Nello Cristianini | journal paper | 2012 |
| Leveraging Noisy Online Databases for Use in Chord Recognition | Matt McVicar, Ni Yizhao, Raul Santos-Rodriguez, Tijl De Bie | conference paper | 2011 |
| Using Online Chord Databases to Enhance Chord Recognition | Matt McVicar, Ni Yizhao, Raul Santos-Rodriguez, Tijl De Bie | journal paper | 2011 |
| Mining the Correlation between Lyrical and Audio Features | Matt McVicar, Tim Freeman, Tijl De Bie | conference paper | 2011 |
| Enhancing Chord Recognition Accuracy using Web Resources | Matt McVicar, Tijl De Bie | conference paper | 2010 |
| Machine Learning with Labeled and Unlabeled Data | Tijl De Bie, Thiago Turchetti Maia, Antonio Braga | conference paper | 2009 |
| Integrating Microarray and Proteomics Data to Predict the Response of Cetuximab in Patients with Rectal Cancer | Anneleen Daemen, Olivier Gevaert, Tijl De Bie, Annelies Debucquoy, Jean-Pascal Machiels, Bart De Moor, Karin Haustermans | conference paper | 2008 |
| Magic Moments for Structured Output Prediction | Elisa Ricci, Tijl De Bie, Nello Cristianini | journal paper | 2008 |
| Learning Performance of a Machine Translation System: a Statistical and Computational Analysis | Marco Turchi, Tijl De Bie, Nello Cristianini | conference paper | 2008 |
| A Metamorphosis of Canonical Correlation Analysis into Multivariate Maximum Margin Learning | Sandor Szedmak, Tijl De Bie, David Hardoon | conference paper | 2007 |
| Deploying SDP for Machine Learning | Tijl De Bie | conference paper | 2007 |
| Learning to Align: a Statistical Approach | Elisa Ricci , Tijl De Bie, Nello Cristianini | conference paper | 2007 |
| Discriminative Sequence Labeling by Z-score Optimization | Elisa Ricci, Tijl De Bie, Nello Cristianini | conference paper | 2007 |
| Modeling Sequence Evolution with Kernel Methods | Margherita Bresco, Marco Turchi, Tijl De Bie, Nello Cristianini | journal paper | 2007 |
| Kernel-Based Data Fusion for Gene Prioritization | Tijl De Bie, Leon-Charles Tranchevent, Liesbeth van Oeffelen, Yves Moreau | journal paper | 2007 |
| The Minimum volume covering ellipsoid estimation in kernel-defined feature spaces | Alexander Dolia, Tijl De Bie, Chris Harris, John Shawe-Taylor, Mike Titterington | conference paper | 2006 |
| Semi-supervised learning using semi-definite programming | Tijl De Bie, Nello Cristianini | book chapter | 2006 |
| Fast SDP relaxations of graph cut clustering, transduction, and other combinatorial problems | Tijl De Bie, Nello Cristianini | journal paper | 2006 |
| Eigenproblems in Pattern Recognition | Tijl De Bie, Nello Cristianini, Roman Rosipal | book chapter | 2005 |
| Semi-supervised learning based on kernel methods and graph cut algorithms | Tijl De Bie | phd thesis | 2005 |
| Kernel methods for exploratory data analysis: a demonstration on text data | Tijl De Bie, Nello Cristianini | conference paper | 2004 |
| Learning from General Label Constraints | Tijl De Bie, Johan Suykens, Bart De Moor | conference paper | 2004 |
| A Statistical Framework for Genomic Data Fusion | Gert Lanckriet, Tijl De Bie, Nello Cristianini, Michael Jordan, William Stafford Noble | journal paper | 2004 |
| Convex Methods for Transduction | Tijl De Bie, Nello Cristianini | conference paper | 2003 |
| On the Regularization of Canonical Correlation Analysis | Tijl De Bie, Bart De Moor | conference paper | 2003 |
| Efficiently Learning the Metric using Side-Information | Tijl De Bie, Michinari Momma, Nello Cristianini | conference paper | 2003 |