This is the personal page of Umberto Picchini, an Associate Professor in Mathematical Statistics at the Department of Mathematical Sciences at Chalmers University of Technology and University of Gothenburg, Sweden. Here is my official university profile. I am also a faculty member of the Chalmers AI Research Centre. Follow @uPicchini
Let’s get in touch to talk about statistical inference (especially Bayesian), likelihood-free methods for models with intractable likelihoods, and Monte Carlo statistical methods such as MCMC and sequential Monte Carlo, and the application of statistical inference in applied problems.
Interested in Bayesian methods? Check out this page!
News
- July 2023: Accepted in Statistics in Medicine, Statistical modeling of diabetic neuropathy: Exploring the dynamics of nerve mortality.
- June 2023: NORDSTAT 2023 was a success! 300 participants, 180 talks and 30 posters. A great pleasure to have been (very much) involved with its organization! Here is a summary article.
- Feb 2023: New paper: JANA: jointly amortized neural approximation of complex Bayesian models.
- Feb 2023: New paper: Mathematical modeling of nerve mortality caused by diabetic neuropathy.
- July 2022: I am the chair of the local organization for NORDSTAT 2023. Check out the event’s page at nordstat2023.org.
- July 2022: I am member of the scientific committee for the BayesComp 2023 satellite workshop “Bayesian computing without exact likelihoods” bayescomp2023.
- 27 June 2022: New paper: Guided sequential ABC schemes for intractable Bayesian models.
- June 2022: giving a talk at ISBA in Montreal on our Sequentially guided synthetic likelihoods paper on 28 June in the 1.30pm session. But there is also a poster here.
- May 2022: published on PLOS Comp. Biology: PEPSDI framework for Bayesian inference for mixed-effects stochastic models.
- 1 April 2022: here are slides from my talk at Maths department in Bristol on our PEPSDI work and generally SDE mixed-effects models.
- 8 February 2022: published on Bayesian Analysis, Sequentially guided MCMC proposals for synthetic likelihoods and correlated synthetic likelihoods.