Hyppää pääsisältöön
In English

Henri Pesonen

apulaisprofessori (Assistant Professor), soveltava tilastotiede
Informaatioteknologian ja viestinnän tiedekunta | Tietotekniikka
Tampereen yliopisto
henri.pesonen [at] tuni.fi (henri[dot]pesonen[at]tuni[dot]fi)
+358503498322
Hervannan kampus
https://orcid.org/0000-0003-4500-2926
https://hpesonen.github.io

Oma esittely

Henri Pesonen is an assistant professor of Applied Statistics at Data Science Research Centre, Tampere University. He leads the Computational Statistics research group that focuses on the development of statistical models, methodology and software to be used by practitioners of various research fields. He graduated as Doctor of Science (Technology) from Tampere University of Technology in 2013, and after his graduation he has worked as a university teacher of statistics at the Turku University, as a post doctoral researcher at the Department of Computer Science at the Aalto university, as a post doctoral fellow and a tenured researcher at Oslo Centre for Biostatistics and Epidemiology before returning to Tampere for his current position. 

 

Osaamisalueiden kuvaus

My main research area is Bayesian simulation-based inference (SBI), where inference methodology is developed for statistical models that can be in theory as complex as we can build in collaboration with the domain experts. The main requirement for SBI is that we can generate, or simulate, synthetic data from the models, that is then used to find approximate posterior distributions of the model parameters. The simulation from the realistic models can be computationally expensive and therefore one of my main research foci has been on making the posterior sampling algorithms more efficient or approximating the expensive stochastic models with probabilistic surrogates. 

I'm also interested in probabilistic programming and developing statistical software tools for practitioners. I'm coordinating the development of the open-source Python package elfi (engine for likelihood-free inference) that is a platform for statistical modeling and simulation-based inference. 

 

Uusimmat julkaisut

Kaikki julkaisut