Paul Blomstedt
PhD, Postdoctoral Researcher
- Office:
-
Room T-A337 in Computer Science Building,
Konemiehentie 2, Otaniemi campus area, Espoo - Postal Address:
-
Aalto University School of Science,
Department of Computer Science,
P.O. Box 15400, FI-00076 Aalto, Finland - Telephone:
- +358 50 4300917
- Email:
- firstname.lastname@aalto.fi
About Me
I'm a postdoctoral researcher in the Probabilistic Machine Learning group at the Department of Computer Science at Aalto University and at Helsinki Institute for Information Technology HIIT. The group is also a part of the Finnish Center for Artificial intelligence (FCAI). I received my PhD in statistics in 2013 from the Department of Mathematics at Åbo Akademi University. Prior to joining my current group, I've been a member of the Bayesian Statistics and Fisheries and Environmental Management groups at the University of Helsinki. In the years 2007-2013, I also worked as a statistician at the National Institute for Health and Welfare (THL).
Much of my current work focuses on developing methods for scalable Bayesian inference, distributed probabilistic modeling and learning from multiple data sources. I have previously done work on predictive modeling, Bayesian clustering and applications of state-space models in population dynamics.
Publications
Here is a list of some recent papers. A more extensive list can be found here.
- Spring 2019: Machine Learning: Advanced Probabilistic Methods, Aalto University
- Spring 2018: Special Course on Latent Variable Modeling and Bayesian Matrix Factorization, Aalto University
- Spring 2018: Machine Learning: Advanced Probabilistic Methods, Aalto University
- Spring 2015: Seminar on Markov chain Monte Carlo, Aalto University
- Spring 2010: Grundkurs i statistisk teori, del 1: Sannolikhetslära, Åbo Akademi University
- Spring 2010: Grundkurs i statistisk teori, del 2: Inferensteori, Åbo Akademi University
- Fall 2009: Statistik 1 för biologer, logopeder och psykologer, Åbo Akademi University
- Fall 2008: Introduction to Statistical Theory, part 1: Probability, Åbo Akademi University
- Fall 2008: Introduction to Statistical Theory, part 2: Inference, Åbo Akademi University
Mesquita, D. Blomstedt, P. and Kaski, S. (2019). Embarrassingly parallel MCMC using deep invertible transformations. Accepted to UAI 2019. [Preprint]
Qin, X., Blomstedt, P. and Kaski, S. (2019). Scalable Bayesian non-linear matrix completion. Accepted to IJCAI 2019.
Blomstedt, P., Mesquita, D., Lintusaari, J., Sivula, T., Corander, J. and Kaski, S. (2019). Meta-analysis of Bayesian analyses. arXiv:1904.0448#[stat.ME]. [Preprint]
Lintusaari, J., Blomstedt, P., Sivula, T., Gutmann, M. U. , Kaski, S. and Corander, J. (2019). Resolving outbreak dynamics using Approximate Bayesian Computation for stochastic birth-death models [version 1; referees: 1 approved with reservations]. Wellcome Open Research, 4(14). [Link] [Code]
Qin, X., Blomstedt, P., Leppäaho, E., Parviainen, P. and Kaski, S. (2019). Distributed Bayesian Matrix Factorization with Limited Communication. To appear in Machine Learning. [Link] [Code]
Vehtari, A., Gelman, A., Sivula, T., Jylänki, P., Tran, D., Sahai, S., Blomstedt, P., Cunningham, J. P., Schiminovich, D. and Robert, C. (2018). Expectation propagation as a way of life: A framework for Bayesian inference on partitioned data. arXiv:1412.4869#[stat.CO]. [Preprint] [Code]
Dutta, R., Blomstedt, P. and Kaski, S. (2016). Bayesian inference in hierarchical models by combining independent posteriors. Technical report. [Link]
Blomstedt, P., Dutta, R., Seth, S., Brazma, A. and Kaski, S. (2016). Modelling-based experiment retrieval: A case study with gene expression clustering. Bioinformatics, 32(9), 1388–1394. [Link] [Code]
Mäntyniemi, S., Whitlock, R., Perälä, T., Blomstedt, P., Vanhatalo, J., Rincón, M.M., Kuparinen, A. Pulkkinen, H. and Kuikka, S. (2015). General state-space population dynamics model for Bayesian stock assessment. ICES Journal of Marince Science, 72(8), 2209–2222. [Link]
Blomstedt, P., Tang, J., Granlund, C. and Corander, J. (2015). A Bayesian predictive model for clustering data of mixed discrete and continuous type. IEEE Transactions on Pattern Analysis and Machine intelligence, 37(3), 489–498. [Link] [Code]
Blomstedt, P. and Corander, J. (2015). Posterior predictive comparisons for the two-sample problem. Communications in Statistics – Theory and Methods, 44(2), 376–389. [Link]