Sander Vandenhaute
@sanderhaute.bsky.social
Computational physics/chemistry, and some ML
prev: PhD @ Ghent University 🇧🇪, AI researcher @ Orbital Materials 🇬🇧
now: postdoc @ Rotskoff group (Stanford University)
prev: PhD @ Ghent University 🇧🇪, AI researcher @ Orbital Materials 🇬🇧
now: postdoc @ Rotskoff group (Stanford University)
April 9, 2025 at 12:31 PM
simple Python API; to drive a single 'master' job which then does everything else!
November 25, 2024 at 6:09 PM
simple Python API; to drive a single 'master' job which then does everything else!
scalable molecular simulation: github.com/molmod/psiflow
scientific: ML potentials, DFT and post-HF calculations, (path-integral) MD, replica exchange, alchemical ΔF , hessians, ...
technical: automated job submission, simple Python, scales to >100 nodes, containerized!
scientific: ML potentials, DFT and post-HF calculations, (path-integral) MD, replica exchange, alchemical ΔF , hessians, ...
technical: automated job submission, simple Python, scales to >100 nodes, containerized!
November 25, 2024 at 6:08 PM
scalable molecular simulation: github.com/molmod/psiflow
scientific: ML potentials, DFT and post-HF calculations, (path-integral) MD, replica exchange, alchemical ΔF , hessians, ...
technical: automated job submission, simple Python, scales to >100 nodes, containerized!
scientific: ML potentials, DFT and post-HF calculations, (path-integral) MD, replica exchange, alchemical ΔF , hessians, ...
technical: automated job submission, simple Python, scales to >100 nodes, containerized!
new SOTA on collective variable learning!
gist? Train classifier in feature space of pretrained GNN to predict 'phase' of an atomic geometry:
CV(A->B) = logit(B) - logit(A)
+data-efficient
+invariant wrt trans/rot/perm
+compatible w foundation models!
gist? Train classifier in feature space of pretrained GNN to predict 'phase' of an atomic geometry:
CV(A->B) = logit(B) - logit(A)
+data-efficient
+invariant wrt trans/rot/perm
+compatible w foundation models!
April 8, 2024 at 1:00 PM
new SOTA on collective variable learning!
gist? Train classifier in feature space of pretrained GNN to predict 'phase' of an atomic geometry:
CV(A->B) = logit(B) - logit(A)
+data-efficient
+invariant wrt trans/rot/perm
+compatible w foundation models!
gist? Train classifier in feature space of pretrained GNN to predict 'phase' of an atomic geometry:
CV(A->B) = logit(B) - logit(A)
+data-efficient
+invariant wrt trans/rot/perm
+compatible w foundation models!