AI4Climate seminar: Machine Learning for Climate Change and Environmental Sustainability – Claire Monteleoni – 6th of February 11:00CET

The seminar is on February the 6th, at 11:00 (CET) remotely and in person. The in-person meeting will be held in the SCAI conference room (map at the end of the post).

If you like to attend online, the link for the zoom is here:

Despite the scientific consensus on climate change, drastic uncertainties remain. Crucial questions about regional climate trends, changes in extreme events, such as heat waves and mega-storms, and understanding how climate varied in the distant past, must be answered in order to improve predictions, assess impacts and vulnerability, and inform mitigation and sustainable adaptation strategies. Machine learning can help answer such questions and shed light on climate change. I will give an overview of our climate informatics research, focusing on challenges in learning from spatiotemporal data, along with semi- and unsupervised deep learning approaches to studying rare and extreme events, and precipitation and temperature downscaling.

Claire Monteleoni is a Choose France Chair in AI and Directrice de Recherche at INRIA Paris, an Associate Professor in the Department of Computer Science at the University of Colorado Boulder, and the founding Editor in Chief of Environmental Data Science, a Cambridge University Press journal, launched in December 2020. She joined INRIA in 2023 and has previously held positions at University of Paris-Saclay, CNRS, George Washington University, and Columbia University. She completed her PhD and Masters in Computer Science at MIT and was a postdoc at UC San Diego. She holds a Bachelor’s in Earth and Planetary Sciences from Harvard. Her research on machine learning for the study of climate change helped launch the interdisciplinary field of Climate Informatics. She co-founded the International Conference on Climate Informatics, which turns 12 years old in 2023, and has attracted climate scientists and data scientists from over 20 countries and 30 U.S. states. She gave an invited tutorial: Climate Change: Challenges for Machine Learning, at NeurIPS 2014. She currently serves on the NSF Advisory Committee for Environmental Research and Education.

Journal Club of V. Balaji

We are hereby re-transmiting the Journal Club news:

This is an announcement of a ‘Journal Club’ on the subject of ‘machine learning’ (ML) methods and their potential application in Earth System modeling (ESM). We will read together (with some imagination the name of the journal club can be pronounced ‘ensemble’!) some of the key papers in this nascent domain, to give ourselves ideas and inspiration on how to integrate these methods in our work, to find a common language between the ESM and ML disciplines, between theory and practice, between idealized systems and the Earth system.
The rules of the game are: someone picks an article where she would like to lead a discussion. To be clear, ‘leading’ doesn’t mean explaining the whole paper: everyone is supposed to have read the paper beforehand, and the discussion will cover points in the paper that aren’t clear, potential flaws and holes, and ideas on how to continue down the path ourselves.
I will lead the first one-hour session, and hope that others will volunteer to lead future ones! It will be on a monthly schedule, every third Thursday, at 14h.
Thus the first is scheduled for 21 February 2019.  There is a room reserved at Jussieu at METIS (Campus Pierre et Marie Curie – Salle Henry Darcy – 3e étage – couloir 46-56) and a visio as well to allow participation from Paris and Saclay campuses. Details, see below.
We will begin with these two articles:
– Schneider et al 2017: Earth System Modeling 2.0: A Blueprint for Models That Learn From Observations and Targeted High‐Resolution Simulations.
– Bolton and Zanna 2019: Applications of Deep Learning to Ocean Data Inference and Subgrid Parameterization.

Hope to see you on the 21st. To have some idea of the level of interest, please drop me a note… but please don’t ‘reply all’ !
Venkatramani Balaji

Journée thématique IA – Océan/Atmosphère/Climat à Rennes le 6 Février

Campus IMT Atlantique, Rennes, 6 février 2019

lien vers le site web de la journée

NB: l’inscription est gratuite mais obligatoire avant le 25 janvier via le lien suivant (inscription)

Nous avons le plaisir de vous informer de l’organisation d’une journée thématique Intelligence Artificielle & Océan-Atmosphère-Climat, le 6 février 2019 à Rennes sur le campus de l’IMT Atlantique, à l’initiative conjointe de l’action MANU (Méthodes Mathématiques et Numériques) du programme LEFE et du programme PNTS. Le comité d’organisation de cette journée thématique est composé de J. Brajard (Sorbonne Univ., IPSL), R. Fablet (IMT Atlantique, Lab-STICC), J. Le Sommer (CNRS, IGE), L. Terray (CERFACS, CECI) S. Thiria (Sorbonne Univ., IPSL). 

L’objectif de cette journée est de réunir les acteurs qui s’intéressent à l’interface entre les modèles et technologies de l’intelligence artificielle, notamment de l’apprentissage, et les sciences de l’atmosphère, de l’océan et du climat. Nous souhaitons à la fois présenter et discuter le potentiel et les avancées récentes de l’IA pour les domaines Océan-Atmosphère-Climat et échanger sur les actions pertinentes pour animer, soutenir et structurer ces activités à l’échelle nationale.

Le programme de cette journée comprendra :

– des exposés invités par S. Brunton (Prof. Univ. of Washington) et V. Balaji (Prof., Princeton Univ.)

– des exposés oraux (20’ à 25’) de contributions soumises à cet appel (cf. ci-dessous)

– une table-ronde avec des experts et représentants d’agences et instituts : J. Lambin  (CNES), V. Réquena (GENCI) A. Bentami (Ifremer), P. Braconnot (INSU), P. Dandin (Météo France), INRIA (Marc Schoenauer).

Lieu :  Cette journée se déroulera sur le campus de Rennes de l’IMT Atlantique (lien).

Appel à contributions: Nous sollicitons des contributions tant pratiques que théoriques sur le développement et l’application de modèles, stratégies et technologies de l’intelligence artificielle exploitant les sources de donnés disponibles (e.g., simulations numériques, données in situ, observations satellitaires,…) pour des questions thématiques relevant des domaines Océan-Atmosphère-Climat. Les thèmes d’intérêt possibles incluent notamment :

– IA & représentations (stochastiques) des dynamiques géophysiques

– IA & Assimilation de données

– IA & Détection et reconnaissance de patterns et structures géophysiques

– IA & Emulation de dynamiques géophysiques

– IA et extrêmes.