Rolf Nevanlinna Institute - Mathematical Methods of Information Technology: Publications

Division of Mathematical Methods of Information Technology - Publications

Refereed
Non-refereed
Technical Reports
Theses
More Student Work
Other Publications

Refereed Publications: Journals, Proceedings, and Chapters

P. Erästö and L. Holmström. Bayesian analysis of features in a scatter plot with dependent observations and errors in predictors. Submitted to Journal of Statistical Computation and Simulation, 2005.
J Klemelä. Adaptive estimation of the mode of a multivariate density. Journal of Nonparametric Statistics, 2005. In press.
J Klemelä. Algorithms for the manipulation of level sets of nonparametric density estimates. Computational Statistics, 2005. In press.
P. Erästö and L. Holmström. Bayesian multiscale smoothing for making inferences about features in scatter plots. To appear in Computational and Graphical Statistics, 2004. (PostScript, 26 pages, 884578 bytes)
F. Hoti and L. Holmström. A semiparametric density estimation approach to pattern classification. Pattern Recognition, 37(3):409-419, 2004. (PostScript, 24 pages, 1064890 bytes)
Fabian Hoti and L. Holmström. Application of semiparametric density estimation to classification. In Proceedings of the 17th International Conference on Pattern Recognition, ICPR2004, Volume 3, Session 2P.We-i (Classification), Cambridge, United Kingdom, 2004. IEEE Computer Society Press, Los Alamitos, CA.
F. Hoti, A. Tuulio-Henriksson, J. Haukka, T. Partonen, L. Holmström, and J. Lönnqvist. Family-based clusters of cognitive test performance in familial schizophrenia. BMC Psychiatry, http://www.biomedcentral.com/1471-244X/4/20, 4:20, 2004.
J. Klemelä. Complexity penalized support estimation. Journal of Multivariate Analysis, 88:274-297, 2004.
J. Klemelä. Visualization of multivariate density estimates with level set trees. Journal of Computational and Graphical Statistics, 13(3):599-620, 2004.
J. Klemelä and A. B. Tsybakov. Exact constants for pointwise adaptive estimation under the Riesz transform. Probability Theory and Related Fields, 129(3):441-467, 2004.
P. Koistinen, L. Holmström, and Erkki Tomppo. Considerations in using smoothing methodology for small-area estimation in multi-source forest inventory. Submitted to Remote Sensing of Environment, 2004.
Jukka Sarvas, Jaan Praks, Lisa M. Zurk, Petri Koistinen, Martti Hallikainen, Jouni Pulliainen, and Lasse Holmström. A polarimetric forest scattering model and its validation. Submitted to IEEE Transactions on Geoscience and Remote Sensing, 2004.
F. Hoti and L. Holmström. On the estimation error in binned local linear regression. Journal of Nonparametric Statistics, 15(4-5):625-642, 2003. (PostScript, 24 pages, 319246 bytes)
J Klemelä. Lower bounds for the asymptotic minimax risk with spherical data. Journal of Statistical Planning and Inference, 113:113-136, 2003. (PostScript, 32 pages, 369131 bytes)
J. Klemelä. Optimal recovery and statistical estimation in L_p Sobolev classes. Mathematical Methods of Statistics, 12(4):429-453, 2003.
V. Kolehmainen, S. Siltanen, S. Järvenpää, J. P. Kaipio, P. Koistinen, M. Lassas, J. Pirttilä, and E. Somersalo. Statistical inversion for medical X-ray tomography with few radiographs II: Application to dental radiology. Physics in Medicine and Biology, 48(10):1465-1490, 2003.
S. Siltanen, V. Kolehmainen, S. Järvenpää, J. P. Kaipio, P. Koistinen, M. Lassas, J. Pirttilä, and E. Somersalo. Statistical inversion for medical X-ray tomography with few radiographs I: General theory. Physics in Medicine and Biology, 48(10):1437-1463, 2003.
L. Holmström and P. Erästö. Making inferences about past environmental change using smoothing in multiple time scales. Computational Statistics & Data Analysis, 41(2):289-309, 2002. (PostScript, 29 pages, 1329024 bytes)
J. Horowitz, J. Klemelä, and E. Mammen. Optimal estimation in additive regression models. Submitted for publication, 2002.
F.J. Hoti, M.J. Sillanpää, and L. Holmström. A note on estimating the posterior density of a qualitative trait locus from a Markov chain monte carlo sample. Genetic Epidemiology, 22:369-376, 2002.
B. Knuteson, H.E. Miettinen, and L. Holmström. alpha PDE: A new multivariate technique for parameter estimation. Computer Physics Communications, 145(3):351-356, 2002.
J. Klemelä. Multivariate histograms with data-dependent partitions. Submitted for publication, 2001.
J. Klemelä and A. B. Tsybakov. Sharp adaptive estimation of linear functionals. Annals of Statistics, 29:1567-1600, 2001. (PostScript, 41 pages, 442627 bytes)
L. Holmström. The accuracy and the computational complexity of a multivariate binned kernel density estimator. Journal of Multivariate Analysis, 72(2):264-309, 2000.
J. Klemelä. Estimation of densities and derivatives of densities with directional data. Journal of Multivariate Analysis, 73(1):18-40, 2000.
J. Klemelä, S. Klinke, and H. Sofyan. Classification and regression trees. In W. Härdle, Z Hlávka, and S. Klinke, editors, XploRe - Application Guide, pages 281-304. Springer, 2000.
A. Korhola, J. Weckström, L. Holmström, and P. Erästö. A quantitative Holocene climatic record from diatoms in northern Fennoscandia. Quaternary Research, 54:284-294, 2000.
F. Hoti and L. Holmström. Reduced Kernel Regression for Fast Classification. In Leif Arkeryd, Jöran Berg, Philip Brenner, and Rolf Pettersson, editors, Progress in Industrial Mathematics at ECMI 98, pages 405-412. B. G. Teubner Stuttgart - Leipzig, 1999.
J. Klemelä. Asymptotic minimax risk for the white noise model on the sphere. Scandinavian Journal of Statistics, 26:465-473, 1999.
J. Klemelä. Sharp adaptive estimation of quadratic functionals. Prépublication 529, Laboratoire de Probabilités et Modèles Aléatoires, Universités de Paris 6 et Paris 7, 1999. Submitted for publication.
L. Holmström and F. Hoti. Radial basis function classification as computationally efficient kernel regression. In IJCNN '98, Proceedings of the 1998 IEEE International Joint Conference on Neural Networks, Anchorage, Alaska, May 4-9, pages 1305-1310, 1998.
P. Koistinen. Asymptotic theory for regularization: One-dimensional linear case. In M. I. Jordan, M. J. Kearns, and S. A. Solla, editors, Advances in Neural Information Processing Systems 10, pages 294-300, 1998. (PostScript, 7 pages, 156692 bytes)
L. Holmström. The error and the computational complexity of a multivariate binned kernel density estimator. In D.W. Scott, editor, Computing Science and Statistics, 29(1), pages 519-528. Interface Foundation of North America, Inc., Fairfax Station, VA 22039-7460, 1997.
L. Holmström and S.R. Sain. Multivariate discrimination methods for top quark analysis. Technometrics, 39(1):91-99, February 1997.
L. Holmström, P. Koistinen, J. Laaksonen, and E. Oja. Neural and statistical classifiers--taxonomy and two case studies. IEEE Transactions on Neural Networks, 8(1):5-17, 1997.
P. Koistinen. Convergence in noisy training. In J. C. Mason S. W. Ellacott and I. J. Anderson, editors, Mathematics of Neural Networks: Models, Algorithms and Applications, pages 220-224. Kluwer Academic Publishers, 1997.
P. Koistinen. Large sample results for training with noise. In Proc. Measurement '97, Smolenice Castle, Slovak Republik, May 29-31, pages 344-347, 1997.
A. Hämäläinen and L. Holmström. Complexity reduction in probabilistic neural networks. In C. von der Malsburg, W. von Seelen, J.C.Vorbrüggen, and B. Sendhoff, editors, Artificial Neural Networks-ICANN' 96, Proceedings of the 1996 International Conference, Bochum, Germany, pages 65-70, July 1996. Lecture Notes in Computer Science 1112, Springer.
L. Holmström, P. Koistinen, J. Laaksonen, and E. Oja. Neural network and statistical perspectives of classification. In Proceedings of the 13th International Conference on Pattern Recognition, ICPR-96, Vienna, pages IV: 286-290, Los Alamitos, CA, 1996. IEEE Computer Society Press.
A. Hottinen and L.Holmström. Projection pursuit for CDMA communications. In Proceedings of the 30th Annual Conference on Information Sciences and Systems (CISS'96), pages 101-106, New Jersey, March 1996.
A. Hämäläinen. Using genetic algorithm in self-organizing map design. In D.W. Pearson, N.C. Steele, and R.F. Albrecht, editors, Artificial Neural Nets and Genetic Algorithms. Proceedings of the International Conference ICANNGA'95, Alès, France, pages 364-367. Springer-Verlag, 1995.
L. Holmström. Neural networks vs. statistics: A comparison using high-energy physics data. In A. B. Bulsari and S. Kallio, editors, Engineering Applications of Artificial Neural Networks. Proceedings of the International Conference EANN'95, Otaniemi, 21-23 August 1995, Finland, pages 441-444, 1995.
L. Holmström, A. Hottinen, and A. Hämäläinen. Using a self-organizing kernel density estimator for CDMA communications. In A. B. Bulsari and S. Kallio, editors, Engineering Applications of Artificial Neural Networks. Proceedings of the International Conference EANN'95, Otaniemi, 21-23 August 1995, Finland, pages 445-448, 1995.
L. Holmström, S.R. Sain, and H.E. Miettinen. A new multivariate technique for top quark search. Computer Physics Communications, 88:195-210, 1995.
H.E. Miettinen, L. Holmström, and S.R. Sain. Top quark search with probability density estimates and neural networks. In B. Denby and D. Perret-Gallix, editors, New Computing Techniques in Physics Research IV, pages 473-478, Singapore, 1995. World Scientific.
A. Hämäläinen. A measure of disorder for the self-organizing map. In Proceedings of the 1994 IEEE International Conference on Neural Networks, Orlando, Florida, June 28 - July 2, pages 659-664, 1994.
L. Holmström and A. Hämäläinen. The self-organizing reduced kernel density estimator. In Proceedings of the 1993 IEEE International Conference on Neural Networks, San Francisco, California, March 28 - April 1, volume 1, pages 417-421, 1993.
L. Holmström and T. Kohonen. Neural networks. In E. Hyvönen, I. Karanta, and M. Syrjänen, editors, Encyclopaedia of Artificial Intelligence, pages 85-98. Gaudeamus Oy, 1993. In Finnish.
P. Koistinen. Unsupervised formation of feature detectors using residual inputs. In S. Gielen and B. Kappen, editors, ICANN'93: Proceedings of the International Conference on Artificial Neural Networks, pages 219-223. Springer-Verlag, 1993.
L. Holmström and J. Klemelä. Asymptotic bounds for the expected L¹ error of a multivariate kernel density estimator. Journal of Multivariate Analysis, 42(2):245-266, 1992.
L. Holmström and P. Koistinen. Using additive noise in back-propagation training. IEEE Transactions on Neural Networks, 3(1):24-38, January 1992.
P. Koistinen and L. Holmström. Kernel regression and backpropagation training with noise. In J. E. Moody, S. J. Hanson, and R. P. Lippman, editors, Advances in Neural Information Processing Systems 4, pages 1033-1039, San Mateo, CA, 1992. Morgan Kaufmann Publishers.
J. T. Alander, M. Frisk, L. Holmström, A. Hämäläinen, and J. Tuominen. Process error detection using self-organizing feature maps. In T. Kohonen, K. Mäkisara, O. Simula, and J. Kangas, editors, Artificial Neural Networks, volume 2, pages 1229-1232. Elsevier Science Publishers B.V. (North-Holland), 1991.
Jarmo T. Alander, Antti Autere, Lasse Holmström, Peter Holmström, Ari Hämäläinen, and Juha Tuominen. Surface type recognition by a hair sensor using neural network methods. In Erdal Arikan, editor, Proceedings of the 1990 Bilkent International Conference on New Trends in Communication, Control, and Signal Processing (BILCON), volume II, pages 1757-1764, Ankara, 2. - 5. July 1990.
L. Holmström, P. Koistinen, and R. J. Ilmoniemi. Classification of unaveraged evoked cortical magnetic fields. In Proc. IJCNN-90-WASH DC, pages II: 359-362. Lawrence Erlbaum Associates, 1990.

Non-refereed Papers in Conferences and Collections

P. Erästö and L. Holmström. Bayesian analysis of trends in a two-dimensional scatter plot. In COMPSTAT'04 - 16th Symposium of IASC on Computational Statistics. Book of abstracts, page 254, Prague, Czech Republic, 2004. Czech Statistical Society.
P. Erästö and L. Holmström. Bayesian analysis of trends in a two-dimensional scatter plot. In In 20th Nordic Conference on Mathematical Statistics. Abstracts volume, Jyväskylä, Finland, 2004.
P. Erästö and L. Holmström. BSiZer for making Bayesian inferences about features in scatter plots. In 6th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 67th Annual Meeting of the Institute of Mathematical Statistics. Progrmamme, Abstracts and Directory of Participants, pages 115 -- 116, Barcelona, Spain, 2004.
P. Koistinen, L. Holmström, and E. Tomppo. Using local linear smoothing for predicting regional averages in multi-source forest inventory. Presented at 1st Göttingen GIS & Remote Sensing Days, 7-8 October 2004, Göttingen, Germany, 2004.
P. Erästö and L. Holmström. Bayesian SiZer - a tool for inferring significant features in environmental reconstructions. In 9th International Paleolimnology Symposium, Abstracts Volume, Espoo, Finland, 2003.
P. Erästö and L. Holmström. Bayesian SiZer - a tool for parametric data analysis of scatter plots. In Bulletin of the International Statistical Institute 54th Session, Proceedings (CD-ROM), August 13 -- 20, Berlin, Germany, 2003.
P. Erästö and L. Holmström. Bayesian SiZer - a tool for parametric data analysis of scatter plots. In B. Fournier, R. Furrer, T. Gsponer, and E.-M. Restle, editors, Proceedings of the 13th European Young Statisticians Meeting (EYSM'03), Ovronnaz, Switzerland, September 21-26, 2003, 2003.
L. Holmström. Discussion of the invited paper meeting 19: Numerical methods in statistics including iterative methods for non-linear problems. In Bulletin of the International Statistical Institute 54th Session, Proceedings (CD-ROM), August 13 -- 20, Berlin, Germany, 2003. Invited paper.
F. Hoti and L. Holmström. A semiparametric approach to statistical pattern recognition. In Bulletin of the International Statistical Institute 54th Session, Proceedings (CD-ROM), August 13 -- 20, Berlin, Germany, 2003.
L. Holmström and P. Koistinen. Using additive noise in back-propagation training. In Jukka Iivarinen, Samuel Kaski, and Erkki Oja, editors, Neljännesvuosisata Hatutusta: Hahmontunnistustutkimus Suomessa 1977 --2002, pages 285 -- 301. Suomen hahmontunnistustutkimuksen seura ry, Pattern Recognition Society of Finland, 2002. Reprint of a paper earlier published in IEEE Transactions on Neural Networks (1992).
L. Holmström, P. Koistinen, J. Sarvas, E. Tomppo, and L. Zurk. A polarimetric scattering model and a new approach to the estimation of forest parameters. In Jouni Jussila, Tuomo Nygrén, and Väinö Kelhä, editors, The IX Meeting of Finnish National COSPAR and ANTARES Fall Seminar 2002, page 38, Oulu, Finland, 2002.
P. Kemppainen-Kajola. Bayesian estimation of non-stationary Markov random fields with an application on forestry. In Abstracts of the 19th Nordic Conference on Mathematical Statistics, page 95, Stockholm, Sweden, 2002.
P. Erästö, L. Holmström, A. Korhola, and J. Weckström. Sizer - a tool for inferring significant features in environmental reconstructions. In Past Climate Variability Through Europe and Africa, An International Conference. Abstracts, page 79, Centre des Congrès, Aix-en-Provence, France, August 27-31, 2001.
A. Korhola, J. Weckström, K. Vasko, H. T. Toivonen, L. Holmström, and P. Erästö. Holocene climate records from aquatic organisms in Finnish Lapland: Comparison of various models and proxies. In Mikko Lahti, Linda Talve, Sakari Tuhkanen, and Jukka Käyhkö, editors, CLIC, Climate change variability in northern Europe, Climate change symposium, Programme and abstracts, page 63, Turku/Åbo, Finland, June 6-8th, 2001.
M. Sillanpää, F. Hoti, and L. Holmström. Estimating the posterior density of a quantitative trait locus from a Markov chain Monte Carlo sample. In 7th Quantitative Trait Locus Mapping and Marker-Assisted Selection Workshop, page 41, Universidad Politécnica de Valencia, October 19-20th, 2001.
L. Holmström, P. Erästö, P. Koistinen, J. Weckström, and A. Korhola. Using smoothing to reconstruct the Holocene temperature in Lapland. In E. Wegman and Y. Martinez, editors, Computing Science and Statistics, 32. Modeling the Earth's Systems: Physical to Infrastructural. Proceedings of the 32nd Symposium on the Interface, pages 425-437, Fairfax Station, VA, USA, 2000. Interface Foundation of North America, Inc. Invited paper.
L. Holmström, P. Koistinen, F. Hoti, and P. Erästö. Classification of Complex Data. In Year 2000, 5th World Congress of the Bernoulli Society for Mathematical Statistics and Probability and 63rd Meeting of the Institute of Mathematical Statistics. Progrman, Abstracts and Directory of Participants, page 76, Guanajuato, Mexico, 2000. Invited paper.
L. Holmström, F. Hoti, and P. Koistinen. Experiments in polychotomous classification. In Bulletin of the International Statistical Institute, ISI 99, the 52nd Session of the International Statistical Institute, August 10 -- 18, 1999, Helsinki, Finland, Contributed Papers, Tome LVIII, Three Books, Book 2, page 41, 1999.
F. Hoti. Handwritten digit classification via computationally efficient kernel regression. In E. Pantzar, editor, Tiedosta tutkittua. Raportti Tiedon tutkimusohjelman I tutkijaseminaarista 6.5.1999. Reports of the Information Research Programme of the Academy of Finland 3, 1999, pages 215-219, Tampere, 1999. Tampereen yliopisto, Tietoyhteiskunnan tutkimuskeskus.
P. Koistinen and E. Oja. Intelligent processing and analysis of images and speech. In Eero Pantzar, editor, Tiedosta ja tiedon tutkimuksesta, Suomen Akatemian Tiedon tutkimusohjelman raportteja, 2, pages 141-147. Tampereen yliopistopaino, 1999. in Finnish.
A. Hämäläinen. Geneettiset algoritmit neuroverkkojen opetuksessa ja rakenteen suunnittelussa. In E. Hyvönen and J. Seppänen, editors, Keinoelämä--Artificial Life, Säätytalo 12.5.1995, Helsinki, pages 201-206. Suomen Tekoälyseura--Finnish Artificial Intelligence Society, 1995. In Finnish.
A. Hämäläinen. The reduced kernel density estimator. In Ninth European Young Statisticians Meeting, Econometric Institute, Erasmus University, Rotterdam, August 14-18, pages 36-40, 1995.
L. Holmström and S. Sain. Using multivariate discrimination in top quark search. In American Statistical Association, 1995 Proceedings of the Statistical Computing Section, Orlando, Florida, USA, August 13 -- 17, pages 102-107, 1995.
A. Hämäläinen. GA and Neural Networks. In J. T. Alander, editor, Proceedings of the First Finnish Workshop on Genetic Algorithms and their Applications, November 4-5, 1992, chapter 5, pages 74-84. Helsinki University of Technology, TKO-A30, 1993.
J. Klemelä. Minimax optimality of a kernel density estimator. In Proceedings of the 8th European Young Statisticians Meeting, Palanga, Lithuania, September 5 --12, 1993.
P. Koistinen. Unsupervised formation of feature detectors through residual data clustering. In A. Bulsari and B. Saxén, editors, Neural Network Research in Finland, Proceedings of a symposium held at Åbo Akademi, Turku, Finland, 21 January 1993, pages 1-12. Finnish Artificial Intelligence Society, 1993.
L. Holmström and J. Klemelä. Choosing and L¹ optimal smoothing parameter in kernel density estimation. In Proceedings of the Workshop on Symbolic and Numeric Computation, Helsinki May 30 -- 31, Computing Centre, University of Helsinki, Research Reports 16, 1991.

Technical Reports

L.M. Zurk, P. Koistinen, J. Sarvas, and L. Holmström. Electromagnetic scattering model for forest remote sensing. Research Reports A38, Rolf Nevanlinna Institute, 2002. (PostScript, 38 pages, 2039926 bytes) (PDF, 310774 bytes)
L. Holmström and Panu Erästö. Using the SiZer method in Holocene temperature reconstruction. Research Reports A36, Rolf Nevanlinna Institute, August 2001. (PostScript, 38 pages, 2619036 bytes)
J Klemelä. Multivariate histograms with data-dependent partitions, 2001. Submitted for publication.
J. Klemelä and A. B. Tsybakov. Exact constants for pointwise adaptive estimation under the Riesz transform. Prépublication 699, Laboratoire de Probabilités et Modèles Aléatoires, Universités de Paris 6 et Paris 7, 2001. Submitted for publication. (PostScript, 28 pages, 358582 bytes)
P. Koistinen and L. Holmström. Pattern recognition (in Finnish). In K. Auranen, J. Lukkarinen, J. Seppänen, and S. Vänskä, editors, Collection of Exercises in Applied Mathematics, Rolf Nevanlinna Institute, C36, pages 65-84, 2001.
J. Klemelä. Visualisation of multivariate density estimates with density trees. Research Reports A33, Rolf Nevanlinna Institute, 2000.
J. Klemelä, S. Klinke, and H. Sofyan. Classification and regression trees. Discussion Paper 62, Sonderforschungbereich 373, Humboldt Universität, Berlin, 2000.
J. Klemelä. Asymptotic minimax risk in the uniform norm for the white noise model on the sphere. Discussion Paper 21, Sonderforschungbereich 373, Humboldt Universität, Berlin, 1998. (PostScript, 15 pages, 167056 bytes)
B. Knuteson, H. Miettinen, and L. Holmström. Mass Analysis and Parameter Estimation with PDE. DØ Note 3396, Lawrence Berkeley National Laboratory, Berkeley, California, September 8, 1998.
M. Nussbaum and J. Klemelä. Constructive asymptotic equivalence of density estimation and gaussian white noise. Discussion Paper 53, Sonderforschungsbereich 373, Humboldt Universität, Berlin, 1998.
L. Holmström. The error and the computational complexity of a multivariate binned kernel density estimator. Research Reports A17, Rolf Nevanlinna Institute, July 1997. (PostScript, 38 pages, 374303 bytes)
L. Holmström, P. Koistinen, J. Laaksonen, and E. Oja. Comparison of neural and statistical classifiers--theory and practice. Research Reports A13, Rolf Nevanlinna Institute, 1996.
L. Holmström and S. Sain. Searching for the top quark using multivariate density estimates. Technical Report No 93-3, Department of Statistics, Rice University, Houston Texas 77251-1892, December 1993.
P. Koistinen and L. Holmström. A framework for the design of feature detectors by self-organization. Research Reports A10, Rolf Nevanlinna Institute, 1993.
H.E. Miettinen, R. Ou, L. Holmström, and S. Sain. Searching for top with neural nets II. NN versus probability density estimation. DØ Note 1931, Department of Physics, Rice University, Houston, Texas 77251-1892, November 2 1993.
P. Koistinen and L. Holmström. A framework for the design of feature detectors by self-organization: Final report of subtask 1.1. Technical report, Rolf Nevanlinna Institute, 1992. An internal report of the Esprit basic research project ``Selforganisation and analogical Modeling Using Subsymbolic Computation''.
P. Koistinen and L. Holmström. A framework for the design of feature detectors by self-organization: Preliminary report of subtask 1.1. Technical report, Rolf Nevanlinna Institute, 1992. An internal report of the Esprit basic research project ``Selforganisation and analogical Modeling Using Subsymbolic Computation''.
J. T. Alander, M. Frisk, L. Holmström, A. Hämäläinen, and J. Tuominen. Process error detection using self-organizing feature maps. Research Reports A5, Rolf Nevanlinna Institute, University of Helsinki, 1991.
L. Holmström and J. Klemelä. An asymptotic upper bound for the expected L¹ error of a multivariate kernel density estimator. Research Reports A6, Rolf Nevanlinna Institute, 1991.
A. Hottinen. Monte Carlo Integration of the L¹ error of a kernel density estimate. Research Reports C10, Rolf Nevanlinna Institute, 1991. In Finnish.
A. Autere, J. T. Alander, L. Holmström, P. Holmström, A. Hämäläinen, and J. Tuominen. Surface type recognition by a hair sensor. Research Reports A2, Rolf Nevanlinna Institute, University of Helsinki, 1990.
L. Holmström and P. Koistinen. Using additive noise in back-propagation training. Research Reports A3, Rolf Nevanlinna Institute, December 1990.
L. Holmström, P. Koistinen, and R. J. Ilmoniemi. Classification of unaveraged evoked cortical magnetic fields. Research Reports A1, Rolf Nevanlinna Institute, September 1989.
L. Holmström, P. Koistinen, and J. Sarvas. Using pattern recognition and neural networks techniques in the design of a metal detector gate. Internal Reports C5, Rolf Nevanlinna Institute, 1988.

Theses

S. Etelävuori. Wavelets, approximation and some statistical applications. Master's thesis in applied mathematics, University of Helsinki, October 2004.
F. Hoti. Studies in computational statistical methods with applications to pattern recognition and data analysis. Research Reports A41, Rolf Nevanlinna Institute, 2004. Doctoral thesis, University of Helsinki.
J. Koivunen. Tiheysfunktion estimointi Bayes SiZer-menetelmällä (in Finnish). Master's thesis in applied mathematics, University of Helsinki, April 2004.
P. Erästö. Support vector machines - backgrounds and practice. Licentiate's thesis in applied mathematics, University of Helsinki, 2001.
F. Hoti. Kernel regression via binned data. Research Reports C38, Rolf Nevanlinna Institute, February 2001. Licentiate's thesis in applied mathematics, University of Helsinki.
P. Erästö. Reconstruction of lake-water temperature from fossil chironomid assemblages (in Finnish). Research Reports A28, Rolf Nevanlinna Institute, September 1999. Master's thesis in applied mathematics, University of Helsinki.
F. Hoti. Using kernel regression to estimate class posterior probabilities (in Finnish). Research Reports A19, Rolf Nevanlinna Institute, May 1997. Master's thesis in applied mathematics, University of Helsinki.
J. Klemelä. Estimation of densities and functionals of densities with spherical data. Research Reports A16, Rolf Nevanlinna Institute, 1997. Doctoral thesis, University of Helsinki.
P. Tikkanen. Smoothing splines (in Finnish). Master's thesis in applied mathematics, University of Helsinki, 1997.
A. Hämäläinen. Self-organizing map and reduced kernel density estimation. Research Reports A11, Rolf Nevanlinna Institute, 1995. Doctoral thesis, University of Jyväskylä.
P. Koistinen. Convergence of minimization estimators trained under additive noise. Research Reports A12, Rolf Nevanlinna Institute, 1995. Doctoral thesis, Helsinki University of Technology.
A. Suutari. Application of nonparametric density function estimation to the search for the top quark (in Finnish). Master's thesis in applied mathematics, University of Helsinki, 1995.
A. Hämäläinen. The self-organizing map in density estimation (in Finnish). Licentiate's thesis in information technology, University of Jyväskylä, 1992.
A. Hottinen. Projection pursuit density estimation (in Finnish). Master's thesis in applied mathematics, University of Helsinki, 1992.
J. Klemelä. Convergence of the expected L¹ error of a multivariate kernel estimator (in Finnish). Licentiate's thesis in applied mathematics, University of Helsinki, 1992.
A. Hämäläinen. Estimation of power spectra (in Finnish). Dissertation in applied mathematics, University of Jyväskylä, 1990.
A. Hämäläinen. On pattern recognition methods (in Finnish). Master's thesis, University of Jyväskylä, Department of Mathematics, 1989.

Other Work Related to Degree Studies and Summer Trainee Projects (mostly in Finnish)

J. Koivunen. Todennäköisyysmalliin perustuva k-lähinaapuriluokitin. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 2003.
K. Böss. Puun monistaminen. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 2002.
J. Talponen. Statistical inference by non-parametric methods and reconstruction of Holocene climate record from diatoms in northern Fennoscandia. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 2001.
M. Viljanen. Metsien kaukokartoitusta parametrittomilla regressiomenetelmillä. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 2001.
P. Pankka. Funktionaalinen data-analyysi hahmontunnistustehtävässä. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1999.
P. Erästö. Lämpötilan rekonstruointi järvien pieneliölajien suhteellisista osuuksista ei-Bäyesläisillä menetelmillä. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1998.
F. Hoti. Ydinpainoitteinen luokittelija. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1996.
P. Tikkanen. Eräiden luokittelijoiden vertailua käsinkirjoitettujen numeroiden tunnistuksessa. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1995.
P. Pajunen. Lokalisoidut kantafunktiot tiheys- ja regressiofunktion estimoinnissa. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1993.
J. Ranta. EM-algoritmi. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1992.
A. Hottinen. Neuraalisista PCA-algoritmeista. Kesäharjoitustyö, Rolf Nevanlinna -instituutti, 1991.
A. Hämäläinen. MEG-mittausten analyysiohjelmisto. Sovelletun matematiikan harjoitustyö, Jyväskylän yliopisto, 1989.
S. Ahlfors. MEG-mittausten analysointi hahmontunnistusmenetelmillä. Erikoistyö, Teknillinen korkeakoulu, Kylmälaboratorio, 1988.
A. Hottinen. MEG-mittausten luokittelu aliavaruusmenetelmillä. Erikoistyö, Teknillinen korkeakoulu, Informaatiotekniikan laboratorio, 1988.

Other Publications and Manuscripts

L. Holmström. Estimation of functions (in Finnish). Lecture notes, Rolf Nevanlinna Institute, 2002.
P. Koistinen. Statistical pattern recognition (in Finnish). Lecture notes, Rolf Nevanlinna Institute, 2002. (PDF, 625171 bytes)
L. Holmström. Mass analysis and regression. Manuscript, Rolf Nevanlinna Institute, 1995.
L. Holmström. Statistical pattern recognition (in Finnish). Lecture notes, Rolf Nevanlinna Institute, 1994.
A. Hämäläinen. Self-organization based on a measure of disorder. A manuscript, Rolf Nevanlinna Institute, 1993.
P. Koistinen. Introduction to pattern recognition (in Finnish). Lecture Notes D6, Rolf Nevanlinna Institute, 1993.
P. Koistinen. Introduction to time series analysis (in Finnish). Lecture Notes D5, Rolf Nevanlinna Institute, 1993.
L. Holmström. Neural net work at Rolf Nevanlinna Institute. ECMI Nesletter 6, pp. 23-24, Helsinki University Press, October, 1989.
L. Holmström and P. Koistinen. Robot error detection through learning--a scetch of a neural network approach. Manuscript, Rolf Nevanlinna Institute, 1988.