[Forum SIS] Seminar: Stefano Castruccio 7 April (Statistics Seminar @ Ca' Foscari)

[Ilaria Prosdocimi]

Cari colleghi e colleghe,

segnalo il seminario di Mercoledì prossimo (l'elenco dei prossimi seminari
presso il gruppo di Statistica a Ca' Foscari è disponibile alla pagina del
gruppo https://www.unive.it/pag/16818):

Data: 7 Aprile 2021 - ore 15:00
Titolo: Model- and Data-Driven Approximation of Space-Time Systems. A Tale
of Two Approaches
Speaker: Stefano Castruccio (University of Notre Dame)

Il seminario si potrà seguire tramite la piattaforma Zoom:
https://unive.zoom.us/j/82776377762
Meeting ID: 827 7637 7762 - Passcode: SanMarco1

Abstract
In this talk I will discuss two different approaches to approximate
space-time systems. This first one is model-driven and loosely inspired by
physics, assumes that the system is locally diffusive through a stochastic
partial differential equation, and can be efficiently approximated with a
Gaussian Markov random field. This approximation will be used to produce a
stochastic generator of simulated multi-decadal global temperature, thereby
offering a fast alternative to the generation of large climate model
ensembles.
The second approach is instead data-driven, and relies on (deep) neural
networks in time. Instead of traditional machine learning methods aimed at
inferring an extremely large parameter space, we instead rely on an
alternative fast, sparse and computationally efficient echo state network
dynamics on an appropriately dimensionally reduced spatial field. The
additional computational time is then used to produce an ensemble and
probabilistically calibrate the forecast. The approach will be used to
produce air pollution forecasts from a citizen science network in San
Francisco and forecasting wind energy in Saudi Arabia.

This talk will be based on the two following works:

https://projecteuclid.org/journals/annals-of-applied-statistics/volume-14/issue-2/Compression-of-climate-simulations-with-a-nonstationary-global-SpatioTemporal-SPDE/10.1214/20-AOAS1340.short

https://arxiv.org/abs/2102.01141

Cordiali saluti

Ilaria Prosdocimi

----
Ilaria Prosdocimi
Assistant Professor in Statistics
Ca' Foscari University of Venice
Department of Environmental Sciences, Informatics and Statistic
prosdocimi.ilaria a gmail.com
ilaria.prosdocimi a unive.it
-------------- parte successiva --------------
Un allegato HTML è stato rimosso...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20210402/b6ae3612/attachment.html>