BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20251026T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260329T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.30068.field_data.0@www.diag.uniroma1.it
DTSTAMP:20260413T112304Z
CREATED:20260219T000749Z
DESCRIPTION:Abstract:Deep learning thrives on large volumes of data\, yet s
 ome of the most critical scientific questions of our time must be answered
  with remarkably small datasets. Global average temperatures from thermome
 ter records span roughly 150 years\, reliable data on climate-driven migra
 tions cover only the last few decades\, and Antarctic ice cores yield abou
 t a thousand data points spread over 800\,000 years. In all these settings
 \, conventional deep learning strategies are simply not viable. Neural net
 work techniques\, however\, retain a distinctive advantage: their ability 
 to uncover nonlinear relationships between variables\, which traditional d
 ynamical models may fail to capture. This seminar presents a compact revie
 w of climate applications where a neural network tool specifically designe
 d for short datasets has provided new insights\, from global warming attri
 bution and the role of human versus natural forcings\, to the analysis of 
 climate-migration links in the Mediterranean region. The talk will highlig
 ht how these data-driven analyses complement and\, in several cases\, stre
 ngthen the conclusions of physics-based climate models\, offering an indep
 endent line of evidence on the causes and impacts of climate change. A pos
 sible collaborative study on nonlinear causality relationships in a paleoc
 limatic context will also be proposed.Short Bio:Antonello Pasini is Senior
  Researcher and Climate Physicist at the Institute of Atmospheric Pollutio
 n Research of the Italian National Research Council (CNR-IIA) in Rome\, an
 d teaches Physics of Climate at the University of Roma Tre. His research f
 ocuses on developing and applying mathematical models\, in particular arti
 ficial neural networks\, to identify the causes of climate change at globa
 l and regional scales and to study its impacts on territories\, ecosystems
  and societies. He has published extensively in international journals\, i
 ncluding Scientific Reports (Nature group)\, and edited the volume 'Artifi
 cial Intelligence Methods in the Environmental Sciences' (Springer). He ha
 s also served as Vice-President of the Italian Society for Climate Science
 s. A committed science communicator\, he authored several popular books\, 
 among which 'Effetto serra\, effetto guerra' (ChiareLettere\, 2017\, with 
 G. Mastrojeni)\, 'L'equazione dei disastri' (Codice\, 2020) and 'La sfida 
 climatica' (Codice). His blog 'Il Kyoto fisso'\, hosted by Le Scienze (the
  Italian edition of Scientific American)\, was awarded the Italian Nationa
 l Prize for Science Communication in 2016.
DTSTART;TZID=Europe/Paris:20260309T111500
DTEND;TZID=Europe/Paris:20260309T111500
LAST-MODIFIED:20260219T073722Z
LOCATION:B101
SUMMARY:When Deep Learning Falls Short: Neural Networks for Small Datasets 
 in Climate Science - Antonello Pasini
URL;TYPE=URI:http://www.diag.uniroma1.it/node/30068
END:VEVENT
END:VCALENDAR