Umberto Picchini
Universitetslektor
Avdelningen för tillämpad matematik ochOm Umberto Picchini
Jag är intresserad av statistisk inferens för stokastisk modellering, och särskilt Bayesianska beräkningsmetoder. Till exempel är jag intresserad av MCMC, sekvensiella Monte Carlo-metoder (partikelfilter) och i synnerhet "likelihood-fria" metoder, såsom Approximate Bayesian Computation (ABC). Jag har ett särskilt intresse för stokastisk modellering (till exempel stokastiska differentialekvationer) och tillämpningar inom biomedicin.
-
JANA: Jointly Amortized Neural Approximation of Complex Bayesian
Models
Stefan Radev, Marvin Schmitt, Valentin Pratz, Umberto Picchini, Ullrich Koethe, Paul Buerkner
The 39th Conference on Uncertainty in Artificial Intelligence - 2023 -
Statistical modeling of diabetic neuropathy: Exploring the dynamics of nerve
mortality
Konstantinos Konstantionu, Farnaz Ghorbanpour, Umberto Picchini, Adam Loavenbruck, Aila Särkkä
Statistics in Medicine - 2023 -
Scalable and flexible inference framework for stochastic dynamic single-cell
models
Sebastian Persson, Niek Welkenhuysen, Sviatlana Shashkova, Samuel Wiqvist, Patrick Reith, Gregor W. Schmidt, Umberto Picchini, Marija Cvijovic
PLoS Computational Biology - 2022 -
Sequentially guided MCMC proposals for synthetic likelihoods and correlated synthetic
likelihoods
Umberto Picchini, Umberto Simola, Jukka Corander
Bayesian Analysis - 2022 -
Sequential neural posterior and likelihood
approximation
Samuel Wiqvist, Jes Frellsen, Umberto Picchini
2021 -
Efficient inference for stochastic differential equation mixed-effects models using correlated particle pseudo-marginal
algorithms
S Wiqvist, A Golightly, A.T. McLean, Umberto Picchini
Computational Statistics & Data Analysis - 2021 -
Partially Exchangeable Networks and architectures for learning summary statistics in Approximate Bayesian
Computation
Samuel Wiqvist, Pierre-Alexandre Mattei, Umberto Picchini, Jes Frellsen
Proceedings of the 36th International Conference on Machine Learning - 2019 -
Bayesian inference for stochastic differential equation mixed effects models of a tumor xenography
study
Umberto Picchini, Julie Lyng Forman
Journal of the Royal Statistic Society, Series C: Applied Statistics - 2019 -
Accelerating delayed-acceptance Markov chain Monte Carlo
algorithms
Samuel Wiqvist, Umberto Picchini, Julie Lyng Forman, Kresten Lindorff-Larsen, Wouter Boomsma
2019