Chemical reaction mechanisms with eXplainable Machine Learning

Deep learning systems can predict properties of chemical substances but do so in a purely mathematical way. As a consequence, the user does not get any feedback on what are the essential properties a chemical must have to exhibit a certain property. This hampers getting any insight in to the relationship between molecular structure and the properties. In this project, the aim is to transform mathematical vectors that result from deep learning back to molecular properties that chemists understand and can use to their benefit to develop new molecules.

Faculty Supervisor:

Stijn De Baerdemacker

Student:

Partner:

Ghent University

Discipline:

Physics

Sector:

Education

University:

University of New Brunswick

Program:

Globalink Research Award

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Government

International

Load More

International

Load More