Research Focus

MACHINE LEARNING DENSITY FUNCTIONALS

In progress ...

Predicting Chemical Reactivity

In progress ...

MachinE LEARNING Dispersion INTERACTIONS

The popularity of semiempirical dispersion corrections models may be attributed to (i) their solid performance accross the periodic table, (ii) their availability for many electronic-structure approximations, (iii) their incorporation in various quantum-chemical software packages, and (iv) their very low computational complexity. However, because those models are global with respect to their empirical parameters, they are not flexible enough to correctly describe dispersion interactions of arbitrary (supra-)molecular systems. We employed Gaussian process (GP) regression to adjust for systematic errors in Grimme-type dispersion corrections introducing the associated, statistically improved D3-GP model.

Please check this page for other research interests of mine that I do currently not focus on.