Annonce

Les commentaires sont clos.

Hybrid Deep Learning and Physical approaches. Application to energy transition: improving the sustainability of energy systems

13 Juin 2023


Catégorie : Doctorant


Abstract : Modeling complex physical system is a fundamental task in a wide range of scientific domains. Physics-based models are reliable, interpretable but may suffer from different drawbacks. Hybrid models, exploiting both physical priors and deep learning ability to model data appears as a promising direction to better solve scientific problems. The objective of this PhD proposal is to explore the development of hybrid physics-machine learning models by exploring fundamental aspects and applicative ones in the context of multi-scale energy systems. This work is part of a multi-disciplinary project aimed at developing machine learning solutions for energy transition and renewal. This PhD proposal will focus on the Machine Learning aspects with the development of methodological and theoretical contributions.

 

Contact: Patrick Gallinari, patrick.gallinari@sorbonne-universite.fr, Nicolas Thome, nicolas.thome@sorbonne-universite.fr
Location: Sorbonne Université, Pierre et Marie Curie Campus, 4 Place Jussieu, Paris, Fr
Candidate profile: Master degree in computer science or applied mathematics, Engineering school. Background and experience in machine learning. Good technical skills in programming.
How to apply: please send a cv, motivation letter, grades obtained in master, recommendation letters when possible to patrick.gallinari@sorbonne-universite.fr, nicolas.thome@sorbonne-universite.fr
Start date: October/November 2023
Keywords: deep learning, physics-based deep learning, energy transition systems
Full description: pages.isir.upmc.fr/gallinari/wp-content/uploads/sites/14/2023/06/2023-06-06-PhD-position-description-Hybrid-models-Energy-Systems.pdf