Artificial intelligence approaches are currently investigated in a plethora of research problems that range from molecular reaction mechanisms to high-throughput screening of functional materials. Our work involves the development of novel methods and systematic machine learning driven multiscale simulation schemes that will enable a deeper understanding of structure-property-processing relations in complex materials and facilitate a successful and more efficient molecular design of high performance materials.

The broad spectra of length- and time-scales present in complex chemical systems necessitate the implementation of hierarchical multi-scale approaches. Within this scope, the development of appropriate coarse-graining schemes is often necessary. These schemes involve the substitution of groups of atoms by single interaction sites, thereby reducing the number of the system’s degrees of freedom, while at the same time maintaining the ones that are important for the description of the mechanisms or processes under study. Coarse-graining approaches involve the mapping from the atomistic to a coarse-grained (CG) level and equilibration at the CG level and reverse mapping back to the atomistic representation. Systematic machine learning driven multiscale simulation schemes are currently developed for the determination of the CG mapping and of the CG interaction potentials.

This activity is partially funded by ML-MULTIMEM project.