Inverse Problems and Statistics

Docent Lassi Roininen

Lassi Roininen


Academy of Finland postdoctoral researcher
Inverse problems


I work on a broad spectrum from the fundamental mathematical inverse problems theory to applications in near-space remote sensing and subsurface imaging. I collaborate with high-level international research groups both in academia and industry. My research highlight is the development of the methodology of discretisation-invariant and computationally feasible priors for Bayesian inversion of function-valued unknowns. Applications include e.g. tomography (ionospheric, electrical impedance, X-ray) and radar pulse-compression coding and analysis methods. 

Research interests

  • Inverse problems
  • Bayesian statistics
  • Tomography and radar technique development
  • Computational methods

Recent Academic Positions

  • 2017 Sep–Current: Academy of Finland Post-doctoral Researcher, University of Oulu, Finland 
  • 2017 Jan–Aug: Post-doctoral Research Associate, Imperial College London, UK 
  • 2016 Mar–Dec: Post-doctoral Research Associate, University of Warwick, UK 
  • 2015 Jun–2016 Feb: Post-doctoral Researcher, Tallinn University of Technology, Estonia 
  • 2008 Jan–2015 May:   Researcher, University of Oulu, Finland 


  • 2017 Docent, Applied Mathematics, University of Oulu, Finland
  • 2015 PhD, Applied Mathematics, University of Oulu, Finland

Academy of Finland projects

I am PI in the following Academy of Finland projects

Hypermodels and stable priors for Bayesian inversion with applications in ionospheric tomography and subsurface imaging 01.09.2017 - 31.08.2020

Probabilistic Deep Learning via Hierarchical Stochastic Partial Differential Equations / Consortium: AO 01.01.2018 - 31.12.2019

Research visits