@matthiaskellner.bsky.social
PhD student in @labcosmo.bsky.social
error plots for the PET-MAD-DOS model on different datasets
Anticipating π§βπ Wei Bin's talk at #psik2025 (noon@roomA), π’ a new #preprint using PET and the MAD dataset to train a universal #ml model for the density of states, giving band gaps for solids, clusters, surfaces and molecules with MAE ~200meV. Go to the talk, or check out arxiv.org/html/2508.17...!
28.08.2025 07:19 β π 3 π 2 π¬ 0 π 0
Read the associated publication: pubs.acs.org/doi/abs/10.1...
Many thanks to our collaborators at the Laboratory of Magnetic Resonance (LRM) at EPFL:
Jacob Brian Holmes, Ruben Rodriguez-Madrid, Florian Viscosi and Lyndon Emsley
Funding:
Swiss National Science Foundation, NCCR MARVEL, ERC Horizon
First, we generate a pool of candidate structures and then select the one whose calculated chemical shieldings best match the experimental measurements. Because shielding calculations quickly become the computational bottleneck, machine-learning (ML) models can substantially reduce the cost.
25.08.2025 08:53 β π 0 π 0 π¬ 1 π 0Solid-state NMR spectroscopy is increasingly used for structure determination of organic solids. NMR-based structure determination differs from conventional diffraction methods:
25.08.2025 08:53 β π 1 π 0 π¬ 1 π 0
We're introducing ShiftML3, a new ShiftML model for chemical shielding predictions in organic solids.
* ShiftML3 predicts full chemical shielding tensors
* DFT accuracy for 1H, 13C, and 15N
* ASE integration
* GPU integration
Code: github.com/lab-cosmo/Sh...
Install from Pypi: pip install shiftml
metatensor logo
metatomic logo
π¨ #machinelearning for #compchem goodies from our π§βπ team incoming! After years of work it's time to share. Go check arxiv.org/abs/2508.15704 and/or metatensor.org to learn about #metatensor and #metatomic. What they are, what they do, why you should use them for all of your atomistic ML projects π.
22.08.2025 07:40 β π 10 π 7 π¬ 1 π 2
π DFT-accurate, with built-in uncertainty quantification, providing chemical shielding anisotropy - ShiftML3.0 has it all! Building on a successful @nccr-marvel.bsky.social-funded collaboration with LRMπ§²βοΈ, it just landed on the arXiv arxiv.org/html/2506.13... and on pypi pypi.org/project/shif...
17.06.2025 13:18 β π 18 π 7 π¬ 1 π 0
A schematic of the functioning of a ML/QM hybrid framework
When you combine #machinelearning and #compchem, you need to start worrying at the QM details within your ML architecture. We use our indirect Hamiltonian framework and pySCFAD to explore the enormous design space arxiv.org/abs/2504.01187
03.04.2025 21:30 β π 14 π 4 π¬ 2 π 0
Polar plot showing the errors of several machine-learning potential of different test sets. Smaller is better here!
Plots showing the evaluation time per atom for several machine-learning potentials as a function of the number of atoms in a simulation. Smaller is better
π’ PET-MAD has just landed! π’ What if I told you that you can match & improve the accuracy of other "universal" #machinelearning potentials training on fewer than 100k atomic structures? And be *faster* with an unconstrained architecture that is conservative with tiny symmetry breaking? Sounds like π§βπ
19.03.2025 07:23 β π 28 π 9 π¬ 1 π 3
Header of the webpage showing the title ("Atomistic Water model for MD") and the authors (Philip Loche, Marcel Langer, Michele Ceriotti)
Happy to share a new #cookbook recipe that shocases several new software developments in the lab, using the good ole' QTIP4P/f water model as an example. atomistic-cookbook.org/examples/wat.... TL;DR - you can now build torch-based interatomic potentials, export them and use them wherever you like!
28.02.2025 12:58 β π 12 π 5 π¬ 1 π 0Feeling a bit lonely here ...
15.11.2024 11:38 β π 0 π 0 π¬ 0 π 0
