Huyen T.T. Do
(DO Thi Thanh Huyen - Đỗ Thị Thanh Huyền)|
|
Research Assistant 7, route de Drize, Batelle Batiment A Email: Huyen.Do
at unige.ch |
NewsFeb 2013: I will join Microsoft Research Cambridge for an internship. Oct 22, 2012: I just defended my thesis, entitled "A unified framework for Support Vector Machines, Multiple Kernel Learning and Metric Learning". Thesis committee: Internal: Dr. Alexandros Kalousis, Dr. Melanie Hilario, Prof. Christian Pellegrini (University of Geneva, Switzerland) External: Prof. Massimiliano Pontil (University College London, UK), Prof. Ulf Brefeld (Technical University of Damstadt, Germany) |
About me
I have been doing my PhD at AI group, Computer Science Department, University of Geneva since 2008 and have graduated recently (in Oct 2012). My current research interests include, but are not limited to: Kernel methods, including Support Vector Machines, Multiple Kernel Learning, learning kernels, and Transfer Learning, such as multi-task learning and cross domain transfer. I am also interested in discovering relations among machine learning algorithms.
Education
University of Geneva, Switzerland
PhD candidate in Computer Science, 2008-2012.
Thesis supervisors: Alexandros Kalousis (main supervisor)
Melanie Hilario (co-supervisor)
Prof. Christian Pellegrini (director)Saint Petersburg State University, Russia
BSc in Information Technology, 2003-2007.
Teaching Assistant
Master course: Methods and Heuristics for Learning and Optimization (Méthodes et Heuristiques d'Optimisation et d'Apprentissage), Fall 2011, Spring 2012.
Master course: Data mining, Fall 2009, Spring 2010 .
Bachelor course: Software and Computer Networks (Logiciels et Réseaux Informatiques), Spring 2009.
Projects
DropTop (Knowledge Discovery in the Life Sciences)
My task was to create and to analyze survival models of clinical data, which were patients of recurrence or non-recurrence tumor, to discover gene patterns associated with tumor-recurrence. Data came from multiple sources of DNA, protein and mutation of a very limited number of samples (patients).e-Lico (Machine Learning and Data Mining)
I participate in developing DMOP - a Data Mining Optimization ontology, and in ontology-based meta learning experiments.
Publications
Convex formulations of radius-margin based Support Vector Machines. Accepted to International Conference on Machine Learning - ICML 2013. (pdf) ( Appendix)
Huyen Do and Alexandros Kalousis.A unified framework for Support Vector Machines, Multiple Kernel Learning and Metric Learning. PhD Thesis 2012 (link) (pdf)
Huyen Do.A metric learning perspective of SVM: on the relation of SVM and LMNN. In JMLR W&C: Proceedings of Fifteenth International Conference on Artificial Intelligence and Statistics - AISTATS 2012. (pdf) (link) (poster)
Huyen Do, Alexandros Kalousis, Jun Wang and Adam Woznica.On the relation of SVM and LMNN. To appear in NIPS 2011 - workshop on Beyond Mahalanobis: Supervised Large-Scale Learning of Similarity. (link)
Huyen Do, Alexandros Kalousis, Jun Wang and Adam Woznica.Metric learning with multiple kernels. In Proceedings of Neural Information Processing Systems - NIPS 2011 (link) .Jun Wang, Huyen Do, Adam Woznica and Alexandros Kalousis.
Ontology based meta-mining of knowledge discovery workflows. Book chapter in Meta-Learning in Computational Intelligence. Springer 2011. (link)
Melanie Hilario, Phong Nguyen, Huyen Do, Adam Woznica and Alexandros Kalousis.Margin and Radius Based Multiple Kernel Learning. In Proceedings of the European Conference on Machine Learning (ECML-2009), Bled, Slovenia, 2009. (pdf) (link)
Huyen Do, Alexandros Kalousis, Adam Woznica and Melanie Hilario.Feature weighting using margin and radius based error bound optimization in SVMs. In Proceedings of the European Conference on Machine Learning (ECML-2009), Bled, Slovenia, 2009. (pdf) (link)
Huyen Do, Alexandros Kalousis and Melanie Hilario.
Papers in preparation:
On kernel and metric learning (in preparation for JMLR 2012)
Huyen Do, Alexandros Kalousis, Jun Wang and Adam Woznica.