arXiv subjects

Nikolaj Rønne

Publications and source records attributed to Nikolaj Rønne.

8 recordsLinked to original sources

arXiv2026

HIP: Hessian Interatomic Potentials without derivatives

Molecular Hessians, the second derivatives of the potential energy, are fundamental to many workflows in computational chemistry. Usually, accurate Hessians are computationally expensive to calculate and scale poorly with system size, whether computed using quantum chemistry methods or machine-learning interatomic potentials (MLIPs). In this work, we introduce Hessian interatomic potentials (HIPs), a deep learning model that directly predicts Hessians without relying on automatic differentiation or finite differences. To do so, we construct SE(3)-equivariant, symmetric Hessians from irreducible representation (irrep) features up to degree $l$=2, computed by a graph neural network. HIP Hessians are one to two orders of magnitude faster, more accurate, more memory efficient, easier to train, and exhibit more favourable scaling with system size. We validate our predictions across a wide range of downstream tasks, demonstrating consistently superior performance in transition state search, geometry optimization, zero-point energy corrections, and vibrational analysis. We open-source the HIP code and model weights.

cs.LG↗

arXiv2025

GO-Diff: Data-free and amortized global structure optimization

We introduce GO-Diff, a diffusion-based method for global structure optimization that learns to directly sample low-energy atomic configurations without requiring prior data or explicit relaxation. GO-Diff is trained from scratch using a Boltzmann-weighted score-matching loss, leveraging only the known energy function to guide generation toward thermodynamically favorable regions. The method operates in a two-stage loop of self-sampling and model refinement, progressively improving its ability to target low-energy structures. Compared to traditional optimization pipelines, GO-Diff achieves competitive results with significantly fewer energy evaluations. Moreover, by reusing pretrained models across related systems, GO-Diff supports amortized optimization - enabling faster convergence on new tasks without retraining from scratch.

physics.comp-ph↗

arXiv2025

Atomistic Generative Diffusion for Materials Modeling

We present a generative modeling framework for atomistic systems that combines score-based diffusion for atomic positions with a novel continuous-time discrete diffusion process for atomic types. This approach enables flexible and physically grounded generation of atomic structures across chemical and structural domains. Applied to metallic clusters and two-dimensional materials using the QCD and C2DB datasets, our models achieve strong performance in fidelity and diversity, evaluated using precision-recall metrics against synthetic baselines. We demonstrate atomic type interpolation for generating bimetallic clusters beyond the training distribution, and use classifier-free guidance to steer sampling toward specific crystallographic symmetries in two-dimensional materials. These capabilities are implemented in Atomistic Generative Diffusion (AGeDi), an open-source, extensible software package for atomistic generative diffusion modeling.

physics.comp-ph↗

arXiv2024

Augmentation of Universal Potentials for Broad Applications

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\cite{deng2023chgnet} has the potential to succeed out of the box, but can also fail significantly in predicting the ground state configuration. We demonstrate that via fine-tuning or a $Δ$-learning approach it is possible to augment the overall performance of universal potentials for specific cluster and surface systems. We utilize this to investigate and explain experimentally observed defects in the Ag(111)-O surface reconstruction and explain the mechanics behind its formation.

cond-mat.mtrl-sci↗

arXiv2024

Efficient ensemble uncertainty estimation in Gaussian Processes Regression

Reliable uncertainty measures are required when using data based machine learning interatomic potentials (MLIPs) for atomistic simulations. In this work, we propose for sparse Gaussian Process Regression type MLIP a stochastic uncertainty measure akin to the query-by-committee approach often used in conjunction with neural network based MLIPs. The uncertainty measure is coined \textit{"label noise"} ensemble uncertainty as it emerges from adding noise to the energy labels in the training data. We find that this method of calculating an ensemble uncertainty is as well calibrated as the one obtained from the closed-form expression for the posterior variance when the sparse GPR is treated as a projected process. Comparing the two methods, our proposed ensemble uncertainty is, however, faster to evaluate than the closed-form expression. Finally, we demonstrate that the proposed uncertainty measure acts better to support a Bayesian search for optimal structure of Au$_{20}$ clusters.

physics.comp-ph↗

arXiv2024

Generative diffusion model for surface structure discovery

We present a generative diffusion model specifically tailored to the discovery of surface structures. The generative model takes into account substrate registry and periodicity by including masked atoms and $z$-directional confinement. Using a rotational equivariant neural network architecture, we design a method that trains a denoiser-network for diffusion alongside a force-field for guided sampling of low-energy surface phases. An effective data-augmentation scheme for training the denoiser-network is proposed to scale generation far beyond structure sizes represented in the training data. We showcase the generative model by investigating multiple surface systems and propose an atomistic structure model for a previously unknown silver-oxide domain-boundary of unprecedented size.

physics.comp-ph↗

arXiv2022

Atomistic Global Optimization X: A Python package for optimization of atomistic structures

Modelling and understanding properties of materials from first principles require knowledge of the underlying atomistic structure. This entails knowing the individual identity and position of all involved atoms. Obtaining such information for macro-molecules, nano-particles, clusters, and for the surface, interface, and bulk phases of amorphous and solid materials represents a difficult high dimensional global optimization problem. The rise of machine learning techniques in materials science has, however, led to many compelling developments that may speed up such structure searches. The complexity of the new methods have established the necessity for an efficient way of experimenting with and assembling them into global optimization algorithms. In this paper we introduce the Atomistic Global Optimization X (AGOX) framework and code, as a customizable approach to building efficient global optimization algorithms. A modular way of expressing global optimization algorithms is described and modern programming practices are used to enable that modularity in the freely available AGOX python package. Two examples of global optimization problems are analyzed: One that is computationally inexpensive which is used to showcase that AGOX enables the expression of multiple global optimization algorithms. As the other example, AGOX is used for solving a complex atomistic optimization problem for a metal-nitride nano-cluster embedded in a graphene sheet as described at the density functional theory (DFT) level.

physics.chem-ph↗

arXiv2022

Atomistic structure search using local surrogate mode

We describe a local surrogate model for use in conjunction with global structure search methods. The model follows the Gaussian approximation potential (GAP) formalism and is based on a the smooth overlap of atomic positions descriptor with sparsification in terms of a reduced number of local environments using mini-batch $k$-means. The model is implemented in the Atomistic Global Optimization X framework and used as a partial replacement of the local relaxations in basin hopping structure search. The approach is shown to be robust for a wide range of atomistic system including molecules, nano-particles, surface supported clusters and surface thin films. The benefits in a structure search context of a local surrogate model are demonstrated. This includes the ability to transfer learning from smaller systems as well as the possibility to perform concurrent multi-stoichiometry searches.

physics.chem-ph↗