.jpg)
The lecture presented a brief overview on the research done at the Hylleraas Centre in the fields of evolutionary and deep machine learning, and how these two approaches are being combined[1] in the discovery of transition metal complexes. The focus was on multiobjective optimization problems[2], moving from 2D Pareto fronts to 3D Pareto surfaces using genetic algorithms. Further, efforts in implementing a variational autoencoder based on the junction-tree formalism for the unconditional and conditional generation of metal ligands[3] were also discussed, including recent work on the combination of this approach with genetic algorithms.
https://quantchem.weebly.com/the-2024-loumlwdin-symposium.html
References:
[1] Kneiding, H.; Balcells, D. Chem. Sci., 2024, 15, 15522;
[2] Kneiding, H.; Nova, A.; Balcells, D. Nat. Comput. Sci., 2024, 4, 263;
[3] Strandgaard, M.; Linjordet, T.; Kneiding, H.; Burnage, A.; Nova, A.; Jensen, J. H.; Balcells, D. ChemRxiv, 2024, DOI: 10.26434/chemrxiv-2024-mzs7b.
Presentation: 2024.12.06, University of Uppsala, Uppsala, Sweden.