*
Quick Links|Home|Worldwide
Microsoft*
Search for
 
Picture of Joaquin
Joaquin Quiñonero Candela
Researcher

Online Services and Advertising
Machine Learning and Perception
Microsoft Research Cambridge
Interests I am interested in applying Machine Learning techniques to challenging real-world problems with massive amounts of data. The Web nicely offers plenty of these!

Within Machine Learning, I am interested in probabilistic models and in Bayesian inference. In the past, I have worked with Gaussian Process priors and Relevance Vector Machine models, and provided a new unifying view on sparse approximations to Gaussian Processes [paper in pdf].
Publications
  1. Miguel Lázaro-Gredilla, Joaquin Quiñonero-Candela and Aníbal Figueiras-Vidal. Sparse Spectral Sampling Gaussian Processes. 2007, Microsoft Research Technical Report MSR-TR-2007-152.
    [technical report MSR-TR-2007-152]
  2. Joaquin Quiñonero-Candela, Edward Snelson and Oliver Williams. Sensible Priors for Sparse Bayesian Learning. 2007, Microsoft Research Technical Report MSR-TR-2007-121.
    [technical report MSR-TR-2007-121]
  3. Joaquin Quiñonero-Candela, Carl Edward Rasmussen, and Christopher K. I. Williams. Approximation Methods for Gaussian Process Regression. In Leon Bottou, Olivier Chapelle, Dennis DeCoste and Jason Weston, editors, Large Scale Learning Machines, pages 203-223, Cambridge, MA, 2007. MIT Press.
    [book homepage | technical report MSR-TR-2007-124]
  4. Neil D. Lawrence, Anton Schwaighofer and Joaquin Quiñonero-Candela, editors. JMLR Workshop and Conference Proceedings Volume 1: Gaussian Processes in Practice. Journal of Machine Learning Research, 2007.
    [web]
  5. Neil D. Lawrence and Joaquin Quiñonero-Candela. Local distance preservation in the gp-lvm through back constraints. In W. Cohen and A. Moore, editors, Proceedings of the International Conference in Machine Learning, pages 513-520, San Francisco, CA, 2006. Morgan Kauffman.
    [     bib | .pdf ]
  6. Joaquin Quiñonero-Candela, Carl Edward Rasmussen, Fabian Sinz, Olivier Bousquet, and Bernhard Schölkopf. Evaluating predictive uncertainty challenge. In Joaquin Quiñonero-Candela, Ido Dagan, Bernardo Magnini, and Florence D'Alché-Buc, editors, Evaluating Predictive Uncertainty, Visual Object Categorization and Textual Entailment, volume 3944 of Lecture Notes in Computer Science, pages 1-27, Heidelberg, Germany, 2006. Springer.
    [     bib | .pdf ]
  7. Joaquin Quiñonero-Candela, Ido Dagan, Bernardo Magnini, and Florence D'Alché-Buc, editors. Evaluating Predictive Uncertainty, Visual Object Categorization and Textual Entailment, volume 3944 of Lecture Notes in Computer Science, Heidelberg, Germany, 2006. Springer.
    [     bib ]
  8. Joaquin Quiñonero-Candela and Carl Edward Rasmussen. A unifying view of sparse approximate gaussian process regression. Journal of Machine Learning Research, 6:1935-1959, 2005.
    [     bib | .pdf ]
  9. Carl Edward Rasmussen and Joaquin Quiñonero-Candela. Healing the relevance vector machine by augmentation. In L. De Raedt and S. Wrobel, editors, Proceedings of the 22nd International Conference on Machine Learning, pages 689-696, 2005.
    [     bib | .pdf ]
  10. Alexander Zien and Joaquin Quiñonero-Candela. Large margin non-linear embedding. In L. De Raedt and S. Wrobel, editors, Proceedings of the 22nd International Conference on Machine Learning, pages 1065-1072, 2005.
    [     bib | .pdf ]
  11. Joaquin Quiñonero-Candela and Carl Edward Rasmussen. Analysis of some methods for reduced rank gaussian process regression. In Roderick Murray-Smith and Robert Shorten, editors, Switching and Learning in Feedback Systems, volume 3355 of Lecture Notes in Computer Science, pages 98-127, Heidelberg, Germany, January 2005. Springer.
    [     bib | .pdf ]
  12. Joaquin Quiñonero-Candela. Learning with Uncertainty - Gaussian Processes and Relevance Vector Machines. PhD thesis, Technical University of Denmark, Lyngby, Denmark, 2004.
    [     bib | .pdf ]
  13. F. Sinz, J. Quiñonero-Candela, G. H. Bakir, C. E. Rasmussen, and M.O. Franz. Learning depth from stereo. In Carl Edward Rasmussen, Henrich H. Bülthoff, Martin A. Giese, and Bernhard Schölkopf, editors, Proc. 26 DAGM Pattern Recognition Symposium, pages 245-252, Heidelberg, Germany, 2004. Springer.
    [     bib | .pdf ]
  14. Joaquin Quiñonero-Candela and Ole Winther. Incremental gaussian processes. In Suzanna Becker, Sebastian Thrun, and Klaus Obermayer, editors, Advances in Neural Information Processing Systems 15, pages 1001-1008, Cambridge, MA, 2003. The MIT Press.
    [     bib | .pdf ]
  15. Agathe Girard, Carl Edward Rasmussen, Joaquin Quiñonero-Candela, and Roderick Murray-Smith. Gaussian process with uncertain inputs - application to multiple-step ahead time-series forecasting. In Suzanna Becker, Sebastian Thrun, and Klaus Obermayer, editors, Advances in Neural Information Processing Systems 15, pages 529-536, Cambridge, MA, 2003. The MIT Press.
    [     bib | .pdf ]
  16. Joaquin Quiñonero-Candela, Agathe Girard, Jan Larsen, and Carl Edward Rasmussen. Propagation of uncertainty in bayesian kernels models - application to multiple-step ahead forecasting. In Proceedings of the International Conference on Acoustics, Speech and Signal Processing, volume 2, pages 701-704, Piscataway, New Jersey, 2003. IEEE.
    [     bib | .pdf ]
  17. Joaquin Quiñonero-Candela, Agathe Girard, and Carl Edward Rasmussen. Prediction at an uncertain input for gaussian processes and relevance vector machines - application to multiple-step ahead time-series forecasting. Technical Report IMM-2003-18, Technical University of Denmark, Lyngby, Denmark, 2003.
    [     bib | .pdf ]
  18. Joaquin Quiñonero-Candela and Lars Kai Hansen. Time series prediction based on the relevance vector machine with adaptive kernels. In Proceedings of the International Conference on Acoustics, Speech, and Signal Processing, volume 1, pages 985-988, Piscataway, New Jersey, 2002. IEEE.
    [     bib | .pdf ]
   

©2008 Microsoft Corporation. All rights reserved. Terms of Use |Trademarks |Privacy Statement