6 papers
Efficient Grand Canonical Global Optimization with On-the-fly-trained Machine-learning Interatomic Potentials
Jon Eunan Quinlivan Dominguez, Mads-Peter Verner Christiansen, Konstantin M. Neyman +2
The characterization of nanostructured materials under reactive environments is challenging due to the complexity of the structural motifs involved and their chemical transformatio…
Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence
Mads-Peter Verner Christiansen, Bjørk Hammer
Machine learning interatomic potentials have become an indispensable tool for materials science, enabling the study of larger systems and longer timescales. State-of-the-art models…
Active Î-learning with universal potentials for global structure optimization
Joe Pitfield, Mads-Peter Verner Christiansen, Bjørk Hammer
Universal machine learning interatomic potentials (uMLIPs) have recently been formulated and shown to generalize well. When applied out-of-sample, further data collection for impro…
-model correction of Foundation Model based on the models own understanding
Mads-Peter Verner Christiansen, Bjørk Hammer
Foundation models of interatomic potentials, so called universal potentials, may require fine-tuning or residual corrections when applied to specific subclasses of materials. In th…
Augmentation of Universal Potentials for Broad Applications
Joe Pitfield, Florian Brix, Zeyuan Tang +4
Universal potentials open the door for DFT level calculations at a fraction of their cost. We find that for application to systems outside the scope of its training data, CHGNet\ci…
Efficient ensemble uncertainty estimation in Gaussian Processes Regression
Mads-Peter Verner Christiansen, Nikolaj Rønne, Bjørk Hammer
Reliable uncertainty measures are required when using data based machine learning interatomic potentials (MLIPs) for atomistic simulations. In this work, we propose for sparse Gaus…