CV Seppo Pulkkinen

• Radar-based quantitative precipitation estimation and nowcasting

• Statistical modeling of radar and rain gauge data

• Data mining, image processing, computer vision, optimization, numerical algorithms

• PhD degree in applied mathematics, Department of Mathematics and Statistics, University of Turku, 2014

• Master degree in applied mathematics, Department of Mathematics and Statistics, University of Helsinki, 2009

Conference papers:

S. Pulkkinen, J. Koistinen, and T. Kuitunen. Gauge-radar adjustment by using multivariate kernel regression and spatiotemporal Kriging. 8th European Conference on Radar in Meteorology and Hydrology, Garmisch-Partenkirchen, Germany, September 2014

Journal articles:

S. Pulkkinen, M. M. Mäkelä, and N. Karmitsa. A continuation approach to mode-finding of multivariate Gaussian mixtures and kernel density estimates. Journal of Global Optimization, 56(2):459–487, 2013

S. Pulkkinen, M. M. Mäkelä, and N. Karmitsa. A generative model and a generalized trust region Newton method for noise reduction.Computational Optimization and Applications, 57(1):129–165, 2014

S. Pulkkinen. Ridge-based method for finding curvilinear structures from noisy data. Computational Statistics and Data Analysis, 2014.to appear, doi: 10.1016/j.csda.2014.08.007

S. Pulkkinen. Nonlinear kernel density principal component analysis with application to climate data. submitted to Statistics and Computing, conditionally accepted (based on TUCS Technical Report 1091)

S. Pulkkinen. Finding graph embeddings by incremental low-rank semidefinite programming. submitted to Optimization Methods and Software, conditionally accepted (based on TUCS Technical Report 1069)

Master's thesis:

S. Pulkkinen. A Review of Methods for Unconstrained Optimization: Theory, Implementation and Testing, Department of Mathematics and Statistics, University of Helsinki, December, 2008