Papers

 

Active Δ-learning with universal potentials for global structure optimization

Dec 1, 2025
Joe Pitfield, Mads-Peter Verner Christiansen & Bjørk Hammer

Message-Passing Graph Neural Network with node existence

Gradient-based grand canonical optimization enabled by graph neural networks with fractional atomic existence

We show that neural networks allow us to formulate gradients for a property that is traditionally not differentiable and that these gradients allow us to formulate new optimization algorithms.

Nov 18, 2025
Mads-Peter Verner Christiansen & Bjørk Hammer
 

Δ-model correction of foundation model based on the model’s own understanding

May 8, 2025
Mads-Peter Verner Christiansen & Bjørk Hammer
 

Augmentation of Universal Potentials for Broad Applications

Jan 1, 2025
Joe Pitfield, Florian Brix, Zeyuan Tang, Andreas Møller Slavensky, Nikolaj Rønne, Mads-Peter Verner Christiansen & Bjørk Hamme
 

Efficient ensemble uncertainty estimation in Gaussian processes regression

Oct 29, 2024
Mads-Peter Verner Christiansen, Nikolaj Rønne & Bjørk Hammer
 

Cascading symmetry constraint during machine learning-enabled structural search for sulfur-induced Cu (111)-(43 × 43) surface reconstruction

May 1, 2024
Florian Brix, Mads-Peter Verner Christiansen & Bjørk Hammer
 

Generating candidates in global optimization algorithms using complementary energy landscapes

Jun 11, 2023
Andreas Møller Slavensky, Mads-Peter Verner Christiansen, Bjørk Hammer
 

Atomistic structure search using local surrogate model

Nov 7, 2022
Nikolaj Rønne, Mads-Peter Verner Christiansen, Andreas Møller Slavensky, Zeyuan Tang, Florian Brix, Mikkel Elkjær Pedersen, Malthe Kjær Bisbo, Bjørk Hammer
 

Atomistic global optimization X: A Python package for optimization of atomistic structures

Aug 3, 2022
Mads-Peter Verner Christiansen, Nikolaj Rønne & Bjørk Hammer
 

Generating stable molecules using imitation and reinforcement learning

Dec 10, 2021
Søren Ager Meldgaard, Jonas Köhler, Henrik Lund Mortensen, Mads-Peter Verner Christiansen, Frank Noé and Bjørk Hammer
 

Atomistic structure learning algorithm with surrogate energy model relaxation

Aug 11, 2020
Henrik Lund Mortensen, Søren Ager Meldgaard, Malthe Kjær Bisbo, Mads-Peter Verner Christiansen & Bjørk Hammer

Figure 1: (a-b) ASLA algorithm. (c) One-hot encoded ASLA representation, CNN & Q-map output. (d) Adapted ASLA algorithm with a learnable Gaussian-kernel

Gaussian representation for image recognition and reinforcement learning of atomistic structure

Jun 24, 2020
Mads-Peter Verner Christiansen, Henrik Lund Mortensen, Søren Ager Meldgaard & Bjørk Hammer
 

Constructing convex energy landscapes for atomistic structure optimization

Jan 1, 2019
Siva Chiriki, Mads-Peter Verner Christiansen & Bjørk Hammer
No matching items