Mini colloquiums abstract > Machine learning applications in physicsMini-colloque 9 – MC09
Recently, thanks to the power of new computers and free access to reliable computer libraries, Artificial Intelligence (AI) has become an increasingly used tool in various branches of physics. This rise of AI has certainly contributed to what is known as the "fourth paradigm of science", where the processing of massive data makes it possible to establish a relationship between the different variables of a given problem, and thus to gain a better understanding of the physical phenomenon under study, in cases where established models are not valid or even non-existent.
In this context, this mini-colloquium is aimed at SFP members who are interested in the use of machine learning in physics, or who have integrated this tool (deep neural networks or other methods) into their research activities, or who have even developed research focused on machine learning. Three unifying themes have been identified: control and inverse/regression problems and experimental data processing, model discovery and computation, or the physics of learning. The main objective is to exchange feedback on the use of these tools, which in some cases may not be the best solution to the problem at hand. Finally, the mini-colloquium aims to foster interdisciplinary collaborations to further explore the possibilities and limitations of AI in dealing with multi-physics problems. Far from restricting the mini-colloquium to a specific branch of physics, we hope to cover a wide range of research areas in order to optimise the exchange across specialised communities.
This mini-colloquium is supported by GDR Ondes, I-GAIA and BioComp.
Wherever possible, the preferred language for oral and poster presentations will be English.
|
| Online user: 2 | Privacy |
|
