Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LIFAT / RFAI, Université de Tours, France
Durée : 3 years
Contact : nicolas.ragot@univ-tours.fr
Date limite de publication : 2023-06-30

Contexte :
More details here: http://www.rfai.lifat.univ-tours.fr/phd-position-continual-life-long-learning-for-time-series-prediction-in-environmental-sciences/

The JUNON project, driven by the BRGM, is granted from the Centre-Val de Loire region through ARD program (« Ambition Recherche Développement »). The main goal of JUNON is to elaborate digital services through large scale digital twins in order to improve the monitoring, understanding and prediction of environmental resources evolution and phenomena, for a better management of natural resources. Digital twins will allow to virtually reproduce natural processes and phenomena using combination of AI and environmental tools.
JUNON will focus on the elaboration of digital twins concerning quality and quantity of ground waters, as well as emissions of greenhouse gases and pollutants with health effects, at the scale of geographical area corresponding to the North part of the Centre-Val-de-Loire region. These digital twins will rely on geological and meteorological knowledge and data (time series), as well as physic-based models.
The project actors are: BRGM, Université d’Orléans, Université de Tours, CNRS, INRAE, and ATOS and ANTEA companies.

Sujet :
The PhD position will be in the WP4 of Junon, focused on the prediction of quantity of ground waters and/or prediction of ground/air pollutants. Postdocs at the BRGM and LIFAT will have in charge respectively to collect and arrange data (ground waters levels at different locations) and to benchmark predictions with mechanistic models as well as with classical prediction AI tools integrating several sources of information like:
– meteorological data
– spatial information, i.e. geolocalization of sensors and locations of predictions to be made; topological information such as altitude
– integration of knowledge from mechanistic models as well as from expert knowledge (impact of attributes and variables used)
– etc.

The goal of the PhD will be, relying on these data and protocols, to work on new learning algorithms to allow these AI models to learn continuously giving new observed data as a stream. The scientific locks are clearly related to continual learning for Deep Learning prediction models and especially to deal with:
– few shot learning in DL
– drift and anomaly detection,
– plasticity/stability dilemma
– adapting such algorithms to suggested models by postdoc, based on Transformers or Spatio-Temporal Graph Neural networks using heterogeneous data.

Profil du candidat :
Student having a master degree in computer sciences with experiences in deep learning.

To apply, send the following documents by e-mail to nicolas.ragot [at] univ-tours.fr before 20th of June: a CV, a motivation letter, a short description of your experiences in machine/deep learning, references from academics.

Formation et compétences requises :
Master or Engineering degree or equivalent in computer sciences (Machine learning, data sciences) or applied mathematics

– a good experience in data analysis and machine learning (theory and practice of deep learning in python) is required
– experiences/knowledge in time series prediction and environmental science is welcome
– curiosity and ability to communicate (in English at least) and work in collaboration with scientists from other fields
– autonomy and good organization skills

Adresse d’emploi :
The RFAI group (Pattern Recognition and Image Analysis) is part of the LIFAT (EA 6300) computer science lab.
64 avenue Jean Portalis
37200 Tours , FRANCE

Document attaché : 202306051525_Thèse Junon apprentissage continu.pdf