Sergey Kirshner

Contact Information:

Email:

Phone: (765) 496-9543

Office: HAAS 118

Mailing address:
Department of Statistics
Purdue University
250 N. University Street
West Lafayette, IN 47907-2066
USA

I am an Assistant Professor with the Department of Statistics at Purdue University.

Prior to joining Purdue, I was a postdoctoral fellow with Alberta Ingenuity Centre for Machine Learning at the Department of Computing Science at the University of Alberta. Before that, I was a graduate student and then a postdoc at the Donald Bren School of Information and Computer Sciences at the University of California, Irvine in Padhraic Smyth's research group.

Teaching:

STAT 416 Fall 09: Probability

STAT 598L Fall 09: Probabilistic Graphical Models

STAT 416 Spring 09: Probability

Research:

My research interests lie in the area of statistical machine learning, in particular, in modeling of multivariate and temporal dependence, and its applications to real-world problems.

I am looking for interested graduate and undergraduate students to work me with on research projects.

Publications:

Ph.D. Thesis:

Modeling of multivariate time series using hidden Markov models, 202 pages, March 2005 (PDF, ps.gz)

Journal papers:

Analysis of Indian monsoon daily rainfall on subseasonal to multidecadal time scales using a hidden Markov model, A.M. Greene, A.W. Robertson, S. Kirshner, Quarterly Journal of Royal Meteorological Society, 134(633):875-887, April 2008 (Wiley InterScience, preprint)
Graphical models for statistical inference and data assimilation, A.T. Ihler, S. Kirshner, M. Ghil, A.W. Robertson, P. Smyth, Physica D (special issue on Data Assimilation), 230(1-2):72-87, June 2007 (ScienceDirect, preprint)
Subseasonal-to-interdecadal variability of the Australian monsoon over North Queensland, A.W. Robertson, S. Kirshner, P. Smyth, S.P. Charles, B.C. Bates, Quarterly Journal of Royal Meteorological Society, 132(615):519-542, Part B, January 2006 (Wiley InterScience, preprint)
Downscaling of daily rainfall occurrence over Northeast Brazil using a hidden Markov model, A.W. Robertson, S. Kirshner, P. Smyth, Journal of Climate, 17(22):4407-4424, November 2004 (Allen Press, preprint, tech report)

Conference papers:

ICA and ISA using Schweizer-Wolff measure of dependence, S. Kirshner, B. Póczos, Proceedings of the Twenty-Fifth International Conference on Machine Learning, ICML 2008, July 2008 (paper, slides, software)
Learning with tree-averaged densities and distributions, S. Kirshner, Advances in Neural Information Processing Systems, NIPS 2007, December 2007 (preliminary version, slides)
Infinite mixtures of trees, S. Kirshner, P. Smyth, Proceedings of the Twenty-Fourth International Conference on Machine Learning, ICML 2007, June 2007 (paper, poster, slides)
Conditional Chow-Liu tree structures for modeling discrete-valued vector time series, S. Kirshner, P. Smyth, A.W. Robertson, Proceedings of the Twentieth Conference on Uncertainty in Artificial Intelligence, UAI 2004, July 2004 (paper, tech report, slides)
Unsupervised learning from permuted data, S. Kirshner, S. Parise, P. Smyth, Proceedings of the Twentieth International Conference on Machine Learning, ICML 2003, August 2003 (paper, tech report, slides)
Learning to classify galaxy shapes using the EM algorithm, S. Kirshner, I.V. Cadez, P. Smyth, C. Kamath, Advances in Neural Information Processing Systems 15, NIPS 2002, December 2002, MIT Press, 2003. (paper, poster)
Probabilistic model-based detection of bent-double radio galaxies, S. Kirshner, I.V. Cadez, P. Smyth, C. Kamath, E. Cantú-Paz, Proceedings of the Sixteenth International Conference on Pattern Recognition, ICPR 2002, August 2002 (paper, tech report, poster)
Adaptivity in agent-based routing for data networks, D.H. Wolpert, S. Kirshner, C.J. Merz, K. Tumer, Fourth International Conference on Automomous Agents, Agents 2000, Barcelona, Spain, June 2000 (paper)

Talks:

Let it rain: Modeling multivariate rain time series using hidden Markov models, AI Seminar, Department of Computing Science, University of Alberta, Edmonton, Canada, March 2006 (slides)

Software:

MVNHMM Toolbox for modeling multivariate time series with hidden Markov models
SWICA software for independent component analysis

Personal:

Julia
Ellie
Checkers (our furry bum)

Last modified: 26 July 2009