You are here : EUR MANUTECH SLEIGHT > SLEIGHT's research projects

PGML - Research project

Physics-Guided Machine Learning

ABSTRACT

The aim of the PGML project was to combine the strengths of physics and machine learning to improve the modelling of physical phenomena. Currently, deep learning requires large quantities of data to create accurate models, while physics uses rigorous theories with few data. The idea was therefore to integrate theoretical knowledge of physics into machine learning algorithms to create more efficient models with less data. The project has led to a fruitful collaboration between computer science and physics, with each field benefiting from the advances of the other. In particular, it has made it possible to model the functionalisation of surfaces, under certain conditions, using femto-second lasers.





 
PUBLICATIONS




CONFERENCES
  • Eduardo Brandao, Stefan Duffner, Rémi Emonet, Amaury Habrard, François Jacquenet & Marc Sebban, "Is My Neural Net Driven by the {MDL} Principle?", European Conference on Machine Learning and Knowledge Discovery in Databases