Pith. sign in

REVIEW 4 major objections 7 minor 2 cited by

An automated framework for exploring and learning potential-energy surfaces

T0 review · 4 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read An open-source workflow turns random structure searches into ready-to-use machine-learned interatomic potentials.

desk verdict Useful, honest automation paper; the transferability claim needs held-out force metrics. read the letter →

arxiv 2412.16736 v1 pith:UHHK2FVQ submitted 2024-12-21 physics.comp-ph cond-mat.mtrl-sci

Yuanbin Liu , Joe D. Morrow , Christina Ertural , Natascia L. Fragapane , John L. A. Gardner , Aakash A. Naik , Yuxing Zhou , Janine George
show 1 more author
Volker L. Deringer
This is my paper · ORCID
classification physics.comp-phcond-mat.mtrl-sci
keywords automatedmachine-learnedinteratomicpotentialsrandomstructuresearchingGAP-RSSpotential-energysurfaceexplorationhigh-throughputworkflowsatomate2Gaussianapproximationactivelearning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper presents autoplex, an open-source software framework that automates the construction of machine-learned interatomic potentials (MLIPs) from scratch, replacing hand-curated training sets with an iterative loop of random structure searches, DFT single-point evaluations, and model refitting. The central claim is that this largely automated, high-throughput pipeline can produce potentials that are accurate and robust enough for real materials applications, demonstrated on silicon, the titanium–oxygen system, silica, liquid and crystalline water, and phase-change memory materials. If true, it means the most labor-intensive part of building an MLIP—deciding which configurations to compute and curating them—can be handed to the computer. The authors argue that automation of data generation is the next frontier for making ML-driven atomistic simulation a mainstream tool.

What carries the argument

The load-bearing machinery is the iterative GAP-RSS loop: the buildcell code generates random structures under user-chosen constraints, the current MLIP relaxes them, a subset is labelled with single-point DFT, the new data are added to the training set, a new potential is fitted, and the cycle repeats. A Hookean repulsion term keeps atoms from approaching unphysically close during relaxation. The automation layer wraps this loop in workflow-management infrastructure, so thousands of tasks can be submitted and monitored on high-performance computing systems without manual intervention. This loop is what transfers the burden from hand-curated datasets to the random-search parameters.

What would settle it

Train an autoplex GAP-RSS model on a new small-cell random structure set for water and run molecular dynamics at 300 K; if the predicted O–O radial distribution function first peak deviates by more than roughly 0.05 Å from the experimental value, or the hydrogen-bond count falls outside the 3.48–3.84 range, the claim that RSS alone provides transferable robustness would fail.

Watch

Extended reading notes

Core claim

The central claim is that a GAP-RSS-style workflow—randomly generating small-cell structures, relaxing them with a progressively improved Gaussian approximation potential, labelling only with single-point DFT, and refitting—can be implemented as a modular, automated workflow system and still yield potentials that describe not just the training region but also bulk liquids, amorphous phases, and crystallisation dynamics. The demonstrations show energy errors near 0.01 eV per atom for held-out polymorphs, correct polymorph stability ordering at the SCAN level for SiO2, qualitatively correct liquid-water structure and hydrogen-bond counts, and a 350 ps crystallisation simulation of Ge1Sb2Te4. The paper presents this as evidence that automated random searching can serve as a general starting point for MLIP construction, including for materials with little existing domain knowledge.

Load-bearing premise

The load-bearing premise is that small-cell random structure searches, with user-chosen buildcell constraints and a Hookean repulsion term, generate a training distribution diverse enough for the resulting potential to transfer to large bulk liquid and amorphous systems.

Editorial extensions

If this is right

  • MLIP models for a new material system can be created from scratch in an automated run, with the user's main choice being the random-search constraints and the DFT functional.
  • Automated RSS datasets serve as training data not only for GAP but also for other architectures: the paper shows a NequIP model fitted to the same dataset improves extrapolation to ice polymorphs.
  • Because only single-point DFT is needed, higher-rung functionals such as SCAN become affordable for building potentials, enabling qualitatively correct stability ordering where PBE fails.
  • Materials with little domain knowledge, such as In3Sb1Te2, can get a first usable potential without months of hand curation.
  • The workflow integrates with existing high-throughput infrastructure, making iterative MLIP fitting accessible on large HPC systems.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same automation could be extended to multi-element and disordered systems beyond the demonstrated binaries and ternaries, potentially enabling high-throughput screening of MLIPs across many chemistries.
  • Small-cell random searches may under-sample long-ranged or slowly relaxing degrees of freedom, so adding uncertainty-based active learning to the loop could further improve transfer to large amorphous systems.
  • Autoplex-generated RSS datasets could serve as a cheap pre-training or synthetic-data source for foundational MLIP fine-tuning, extending the paper's observation that a NequIP model benefits from the same dataset.
  • The paper's cost estimates for SiO2 suggest that automated MLIP construction could become a routine pre-screening tool in computational materials discovery, not just a specialist technique.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 7 minor

Summary. The paper introduces autoplex, an automated workflow framework that combines random structure searching (RSS) with iterative fitting of machine-learned interatomic potentials, primarily GAP but with interfaces to other architectures such as NequIP, MACE, and ACE. The workflow is demonstrated on elemental silicon, TiO2 and the full Ti-O binary system, SiO2 at PBE and SCAN levels, liquid water and ice using GAP and NequIP, and the phase-change materials Ge1Sb2Te4 and In3Sb1Te2. The central claim is that MLIP fitting can be carried out in a largely automated, high-throughput manner and that the resulting potentials are robust and useful, especially given the ease of creating them from scratch.

Significance. If the claims are substantiated, this is a valuable contribution: it packages iterative RSS-plus-MLIP fitting in an open-source, workflow-based tool that integrates with atomate2 and jobflow, potentially lowering the barrier to building potentials from scratch. The demonstrations span several chemistries and multiple MLIP architectures, and the SiO2/SCAN example shows that the automated pipeline can access higher-rung functionals at moderate cost. Named strengths include the open code with a Zenodo-deposited version, the interoperability with existing workflow ecosystems, and the demonstration that an RSS-generated dataset can be reused with a different architecture (NequIP). The main gap is validation: the robustness claim currently rests on energy errors that are partly in-sample and on qualitative MD comparisons without force-error metrics or explicit held-out tests on the application systems.

major comments (4)
  1. [Results; Fig. 2 and Table 1] The error metrics are reported without an explicit statement of train/test separation. Fig. 2 shows the evolution of energy errors for selected crystalline phases, but it is not stated whether those phases, or rattled versions of them, are included in the training set at each iteration. Table 1 evaluates errors on 10 rattled structures per polymorph; because these are generated from the ground-state structures that the RSS pipeline is designed to find, they are likely inside or very close to the training distribution. As a result, the reported RMSEs do not by themselves establish that the potentials are robust for unseen configurations. Please report held-out errors and state explicitly which structures enter the training set at each stage.
  2. [Describing water; Application to chalcogenide memory materials; Figs. 4 and 5] The MD demonstrations are the strongest evidence for transferability, but no force-error metric is reported anywhere in the manuscript. The liquid-water RDFs and hydrogen-bond analysis (Fig. 4a-c) and the Ge1Sb2Te4 crystallisation simulation (Fig. 5g) are driven by forces, yet the paper only reports energy RMSEs. Please add force RMSEs on held-out configurations representative of the application systems, for example SCAN DFT-MD snapshots of liquid water and AIMD snapshots of amorphous GST/IST. Without such metrics, the conclusion that small-cell RSS data transfer to bulk liquid and amorphous systems in force space is unsubstantiated.
  3. [Describing water; Fig. 4d] Fig. 4d shows that the GAP-RSS model's energies for ice polymorphs are 'highly scattered', and the authors attribute this to low-density phases outside the training distribution. This is a direct counterexample to the general claim that RSS-derived potentials are robust across phases. Please either soften the robustness claim in the abstract and introduction or provide an analysis of when the workflow produces a transferable GAP, for example by detecting extrapolation through uncertainty estimates or active learning. The current text acknowledges the limitation in one sentence but does not reconcile it with the central claim.
  4. [Data availability] The Data Availability statement says that raw data and notebooks 'will be made available via GitHub upon journal publication', so the numerical results in Figs. 2-5 and Tables 1-2 are not currently available to reviewers or readers. Since the contribution is an automated framework that others should be able to reproduce, please deposit the datasets, trained potentials, and plotting notebooks, or at least the train/test splits and error tables, in a public repository with the submitted version.
minor comments (7)
  1. [Describing water] The statement that RSS searches 'typically use small unit cells' should be quantified: please give the cell sizes and number of molecules used in the water RSS runs so the reader can judge how far the liquid-water simulation extrapolates from the training data.
  2. [Describing water; Application to chalcogenide memory materials] The simulation protocols for Figs. 4a-c and 5c-g (ensemble, thermostat, system size, simulation length, and initial configurations) are not given in the Methods; please add these details for reproducibility.
  3. [The autoplex framework] The license is described only as 'permissive'; please specify the exact open-source license in the text or in the repository.
  4. [Table 1] For the entries marked '— a', the footnote states that errors exceed 1 eV/atom and are therefore not meaningful to report; giving approximate numerical values would be more informative for assessing the failure mode.
  5. [Fig. 2 caption] Please clarify whether the 0.001 eV/atom floor is a plotting cutoff or a stopping criterion used during the iterative fitting.
  6. [Methods; ML potentials] Equation (1) defines the total-energy decomposition but the force expression is not given; please state that forces are obtained as derivatives of the total energy and give the relevant implementation detail or reference.
  7. [Results; Ti-O system] The phrase 'standard DFT and GAP fitting settings' should be replaced by the actual settings or a pointer to repository defaults, since these settings are not standard across the community.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the automation claim is evaluated with external benchmarks and the RSS-derived potentials are tested against held-out ice, liquid, and literature data.

full rationale

The paper's central claim is that autoplex automates iterative RSS plus MLIP fitting, which is an engineering contribution rather than a derived quantitative prediction. The training data are generated by the same RSS pipeline, but the most informative evaluations are external to that pipeline: liquid-water RDFs compared to experiment (Fig. 4a to 4c), 52 ice polymorphs from Ref. 80 re-evaluated with SCAN DFT (Fig. 4d), and phase-change RDFs and ring statistics compared to literature AIMD (Fig. 5c to 5f). The SiO2 stability ordering is checked directly against DFT@SCAN energies, and the energy differences in Table 2 are compared with the corresponding DFT functional, so the GAP predictions are not equated with their training labels by construction. Self-citations to earlier GAP-RSS work (Refs. 44 and 48) supply the underlying method, but the present automation and its external tests are independent; no load-bearing reduction to self-citation occurs. The absence of a force-error metric and the extrapolation from small-cell RSS to bulk liquid and amorphous systems are genuine validation gaps and correctness risks, but they are not instances of circular reasoning under the definitions used here.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The framework does not introduce new physical entities. Its free parameters are implementation choices such as RSS search scope, repulsion strength, and GAP hyperparameters, rather than constants fitted to explain a measured phenomenon. The axioms are the standard assumptions of DFT labelling, local-energy decomposition, and random-search coverage that the method inherits from prior literature.

free parameters (3)
  • buildcell RSS parameter sets
    User-chosen composition ranges, cell sizes, and symmetry constraints define the search scope for each system; values are not itemized in the paper.
  • Hookean repulsion parameters
    Strength and cutoff of the repulsive term added in autoplex are not specified in the text; they shape the generated structures.
  • GAP hyperparameters
    Default GAP and SOAP settings are used but not enumerated; the reported accuracies depend on these choices.
assumptions (4)
  • domain assumption Random structure searching with buildcell generates configurations representative of the relevant potential-energy surface.
    Inherited from AIRSS (Refs 41,42) and GAP-RSS (Refs 44,48); the resulting diversity is what makes from-scratch potentials possible.
  • domain assumption DFT single-point energies computed with VASP (PBE, PBEsol, or SCAN) are accurate reference labels for the target potential-energy surface.
    All training labels come from a chosen DFT functional; functional error is inherited and not quantified in this paper.
  • domain assumption Total energy can be decomposed into local atomic-environment contributions (Eq. 1) as required by GAP.
    Valid for short-ranged descriptors such as SOAP used here; this is a standard MLIP assumption, not proved in the paper.
  • domain assumption Iterative retraining of GAP models on gradually growing RSS datasets converges to a useful potential without manual intervention.
    Demonstrated empirically in the paper and prior work; not accompanied by a formal convergence guarantee.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An automated framework for exploring and learning potential-energy surfaces." pith.science (2026). https://pith.science/paper/UHHK2FVQ

@misc{pith2026241216736,
  author       = {Pith},
  title        = {Pith review of: An automated framework for exploring and learning potential-energy surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UHHK2FVQ}},
  note         = {Machine review of arXiv:2412.16736}
}
read the original abstract

Machine learning has become ubiquitous in materials modelling and now routinely enables large-scale atomistic simulations with quantum-mechanical accuracy. However, developing machine-learned interatomic potentials requires high-quality training data, and the manual generation and curation of such data can be a major bottleneck. Here, we introduce an automated framework for the exploration and fitting of potential-energy surfaces, implemented in an openly available software package that we call autoplex (`automatic potential-landscape explorer'). We discuss design choices, particularly the interoperability with existing software architectures, and the ability for the end user to easily use the computational workflows provided. We show wide-ranging capability demonstrations: for the titanium-oxygen system, SiO2, crystalline and liquid water, as well as phase-change memory materials. More generally, our study illustrates how automation can speed up atomistic machine learning -- with a long-term vision of making it a genuine mainstream tool in physics, chemistry, and materials science.

Figures

Figures reproduced from arXiv: 2412.16736 by the authors.

Figure 1
Figure 1. Exploration and fitting of potential-energy surfaces. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Automated exploration of key materials systems. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Accessing higher-rung DFT for SiO2. (a) Energy prediction errors for α-quartz and α￾cristobalite, plotted as a function of computational cost for autoplex runs using the PBE and SCAN functionals, respectively. (b) DFT-computed energies of key SiO2 polymorphs, relative to α-quartz which is set as energy zero. Note that PBE erroneously predicts α-cristobalite as being more stable than α-quartz. (c) Energy difference b… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Liquid water and ice polymorphs. (a) Radial distribution functions (RDFs) of H–O and O–O pairs comparing experimental data (H–O: Ref. 77; O–O: Ref. 78) with GAP-RSS model predic￾tions at 300 K. (b) Same for predictions from a NequIP model, fitted to the same training s…
Figure 5
Figure 5. Figure 5: Structural and dynamic properties of phase-change materials. [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. NepTrain and NepTrainKit: Automated Active Learning and Visualization Toolkit for Neuroevolution Potentials

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A new open-source toolkit automates active learning and dataset management for neuroevolution potentials, with a CsPbI3 case study showing comparable accuracy to hand-curated NEP models.

  2. Full-cycle device-scale simulations of memory materials with a tailored atomic-cluster-expansion potential

    cond-mat.mtrl-sci 2025-02 conditional novelty 6.0 of 10

    A new ACE machine-learned potential enables full-cycle (RESET-SET-RESET) atomistic simulations of Ge-Sb-Te phase-change memory cells at device scale on CPU-only high-performance computers.

Reference graph

Works this paper leans on

110 extracted references · 68 canonical work pages · cited by 2 Pith papers

  1. [1]

    First principles neural network potentials for reactive simulations of large molecular and condensed systems

    Behler, J. First principles neural network potentials for reactive simulations of large molecular and condensed systems. Angew. Chem. Int. Ed. 56, 12828–12840 (2017)

    2017

  2. [2]

    L., Caro, M

    Deringer, V . L., Caro, M. A. & Cs ´anyi, G. Machine learning interatomic potentials as emerging tools for materials science. Adv. Mater .31, 1902765 (2019)

    2019

  3. [3]

    & Clementi, C

    No ´e, F., Tkatchenko, A., M¨uller, K.-R. & Clementi, C. Machine learning for molecular simulation. Annu. Rev. Phys. Chem. 71, 361–390 (2020)

    2020

  4. [4]

    Unke, O. T. et al. Machine learning force fields. Chem. Rev. 121, 10142–10186 (2021)

    2021

  5. [5]

    & Aspuru-Guzik, A

    Friederich, P., H ¨ase, F., Proppe, J. & Aspuru-Guzik, A. Machine-learned potentials for next-generation matter simulations. Nat. Mater .20, 750–761 (2021)

    2021

  6. [6]

    Cheng, B., Mazzola, G., Pickard, C. J. & Ceriotti, M. Evidence for supercritical be- haviour of high-pressure liquid hydrogen. Nature 585, 217–220 (2020). 19

    2020

  7. [7]

    Deringer, V . L. et al. Origins of structural and electronic transitions in disordered sili- con. Nature 589, 59–64 (2021)

    2021

  8. [8]

    Zong, H. et al. Free electron to electride transition in dense liquid potassium.Nat. Phys. 17, 955–960 (2021)

    2021

Show all 110 references
  1. [9]

    Westermayr, J. et al. Deep learning study of tyrosine reveals that roaming can lead to photodamage. Nat. Chem. 14, 914–919 (2022)

    2022

  2. [10]

    Zhang, S. et al. Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential. Nat. Chem. 16, 727–734 (2024)

    2024

  3. [11]

    Wang, T. et al. Ab initio characterization of protein molecular dynamics with AI2BMD. Nature 635, 1019–1027 (2024)

    2024

  4. [12]

    P., Swiler, L

    Thompson, A. P., Swiler, L. P., Trott, C. R., Foiles, S. M. & Tucker, G. J. Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials. J. Comput. Phys. 285, 316–330 (2015)

    2015

  5. [13]

    Moment tensor potentials: A class of systematically improvable inter- atomic potentials

    Shapeev, A. Moment tensor potentials: A class of systematically improvable inter- atomic potentials. Multiscale Model. Simul. 14, 1153–1173 (2016)

    2016

  6. [14]

    Atomic cluster expansion for accurate and transferable interatomic poten- tials

    Drautz, R. Atomic cluster expansion for accurate and transferable interatomic poten- tials. Phys. Rev. B 99, 014104 (2019)

    2019

  7. [15]

    P., Payne, M

    Bart ´ok, A. P., Payne, M. C., Kondor, R. & Cs ´anyi, G. Gaussian approximation poten- tials: The accuracy of quantum mechanics, without the electrons. Phys. Rev. Lett. 104, 136403 (2010)

    2010

  8. [16]

    & Parrinello, M

    Behler, J. & Parrinello, M. Generalized neural-network representation of high- dimensional potential-energy surfaces. Phys. Rev. Lett. 98, 146401 (2007)

    2007

  9. [17]

    Zhang, L., Han, J., Wang, H., Car, R. & E, W. Deep potential molecular dynamics: A scalable model with the accuracy of quantum mechanics. Phys. Rev. Lett. 120, 143001 (2018)

    2018

  10. [18]

    Unke, O. T. & Meuwly, M. Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges. J. Chem. Theory Comput. 15, 3678–3693 (2019)

    2019

  11. [19]

    Batzner, S. et al. E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. 13, 2453 (2022)

    2022

  12. [20]

    P., Simm, G., Ortner, C

    Batatia, I., Kovacs, D. P., Simm, G., Ortner, C. & Csanyi, G. MACE: Higher or- der equivariant message passing neural networks for fast and accurate force fields. In Koyejo, S. et al. (eds.) Advances in Neural Information Processing Systems , vol. 35, 11423–11436 (Curran Assoc...

    2022

  13. [21]

    & Ong, S

    Chen, C. & Ong, S. P. A universal graph deep learning interatomic potential for the periodic table. Nat. Comput. Sci. 2, 718–728 (2022)

    2022

  14. [22]

    Deng, B. et al. CHGNet as a pretrained universal neural network potential for charge- informed atomistic modelling. Nat. Mach. Intell. 5, 1031–1041 (2023)

    2023

  15. [23]

    C., Miceli, G., Caravati, S., Behler, J

    Sosso, G. C., Miceli, G., Caravati, S., Behler, J. & Bernasconi, M. Neural network interatomic potential for the phase change material GeTe. Phys. Rev. B 85, 174103 (2012)

    2012

  16. [24]

    P., Kermode, J., Bernstein, N

    Bart ´ok, A. P., Kermode, J., Bernstein, N. & Cs ´anyi, G. Machine learning a general- purpose interatomic potential for silicon. Phys. Rev. X 8, 041048 (2018)

    2018

  17. [25]

    C., Rohrer, J., Albe, K

    Erhard, L. C., Rohrer, J., Albe, K. & Deringer, V . L. A machine-learned interatomic potential for silica and its relation to empirical models.npj Comput. Mater .8, 90 (2022)

    2022

  18. [26]

    & Deringer, V

    Zhou, Y ., Zhang, W., Ma, E. & Deringer, V . L. Device-scale atomistic modelling of phase-change memory materials. Nat. Electron. 6, 746–754 (2023)

    2023

  19. [27]

    Zhang, D. et al. DPA-2: a large atomic model as a multi-task learner. Preprint at https://arxiv.org/abs/2312.15492

    arXiv

  20. [28]

    Batatia, I. et al. A foundation model for atomistic materials chemistry. Preprint at https://arxiv.org/abs/2401.00096

    arXiv

  21. [29]

    Kaur, H. et al. Data-efficient fine-tuning of foundational models for first- principles quality sublimation enthalpies. Faraday Discuss. Advance Article, DOI: 10.1039/D4FD00107A (2024)

    2024 doi

  22. [30]

    Ben Mahmoud, C., Gardner, J. L. A. & Deringer, V . L. Data as the next challenge in atomistic machine learning. Nat. Comput. Sci. 4, 384–387 (2024)

    2024

  23. [31]

    Kulichenko, M. et al. Data generation for machine learning interatomic potentials and beyond. Chem. Rev. DOI: 10.1021/acs.chemrev.4c00572 (2024)

    2024 doi

  24. [32]

    S., Nebgen, B., Lubbers, N., Isayev, O

    Smith, J. S., Nebgen, B., Lubbers, N., Isayev, O. & Roitberg, A. E. Less is more: Sampling chemical space with active learning. J. Chem. Phys. 148 (2018)

    2018

  25. [33]

    P., Ortner, C

    van der Oord, C., Sachs, M., Kov ´acs, D. P., Ortner, C. & Cs ´anyi, G. Hyperactive learning for data-driven interatomic potentials. npj Comput. Mater .9, 168 (2023)

    2023

  26. [34]

    Kulichenko, M. et al. Uncertainty-driven dynamics for active learning of interatomic potentials. Nat. Comput. Sci. 3, 230–239 (2023)

    2023

  27. [35]

    Vandermause, J. et al. On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events. npj Comput. Mater .6, 20 (2020). 21

    2020

  28. [36]

    & Van Spey- broeck, V

    Vandenhaute, S., Cools-Ceuppens, M., DeKeyser, S., Verstraelen, T. & Van Spey- broeck, V . Machine learning potentials for metal-organic frameworks using an incre- mental learning approach. npj Comput. Mater .9, 19 (2023)

    2023

  29. [37]

    Uncertainty-aware molecular dynamics from Bayesian active learning for phase transformations and thermal transport in SiC

    Xie, Y .et al. Uncertainty-aware molecular dynamics from Bayesian active learning for phase transformations and thermal transport in SiC. npj Comput. Mater .9, 36 (2023)

    2023

  30. [38]

    A., Johnston-Wood, T., Deringer, V

    Young, T. A., Johnston-Wood, T., Deringer, V . L. & Duarte, F. A transferable active- learning strategy for reactive molecular force fields. Chem. Sci. 12, 10944–10955 (2021)

    2021

  31. [39]

    P., Ko, T., Yang, C

    Guan, X., Heindel, J. P., Ko, T., Yang, C. & Head-Gordon, T. Using machine learning to go beyond potential energy surface benchmarking for chemical reactivity.Nat. Comput. Sci. 3, 965–974 (2023)

    2023

  32. [40]

    L., Fako, E., De, S., Sch ¨afer, A

    Schaaf, L. L., Fako, E., De, S., Sch ¨afer, A. & Cs´anyi, G. Accurate energy barriers for catalytic reaction pathways: an automatic training protocol for machine learning force fields. npj Comput. Mater .9, 180 (2023)

    2023

  33. [41]

    Pickard, C. J. & Needs, R. J. High-pressure phases of silane.Phys. Rev. Lett.97, 045504 (2006)

    2006

  34. [42]

    Pickard, C. J. & Needs, R. J. Ab initio random structure searching. J. Phys.: Condens. Matter 23, 053201 (2011)

    2011

  35. [43]

    Jain, A. et al. Commentary: The Materials Project: A materials genome approach to accelerating materials innovation. APL Mater .1, 011002 (2013)

    2013

  36. [44]

    L., Pickard, C

    Deringer, V . L., Pickard, C. J. & Cs ´anyi, G. Data-driven learning of total and local energies in elemental boron. Phys. Rev. Lett. 120, 156001 (2018)

    2018

  37. [45]

    Tong, Q., Xue, L., Lv, J., Wang, Y . & Ma, Y . Accelerating CALYPSO structure predic- tion by data-driven learning of a potential energy surface.Faraday Discuss. 211, 31–43 (2018)

    2018

  38. [46]

    L., Proserpio, D

    Deringer, V . L., Proserpio, D. M., Cs´anyi, G. & Pickard, C. J. Data-driven learning and prediction of inorganic crystal structures. Faraday Discuss. 211, 45–59 (2018)

    2018

  39. [47]

    V ., Tikhonov, E

    Podryabinkin, E. V ., Tikhonov, E. V ., Shapeev, A. V . & Oganov, A. R. Accelerat- ing crystal structure prediction by machine-learning interatomic potentials with active learning. Phys. Rev. B 99, 064114 (2019)

    2019

  40. [48]

    & Deringer, V

    Bernstein, N., Cs ´anyi, G. & Deringer, V . L. De novo exploration and self-guided learn- ing of potential-energy surfaces. npj Comput. Mater .5, 99 (2019). 22

    2019

  41. [49]

    Pickard, C. J. Ephemeral data derived potentials for random structure search. Phys. Rev. B 106, 014102 (2022)

    2022

  42. [50]

    Pickard, C. J. Beyond theory driven discovery: introducing hot random search and da- tum derived structures. Faraday Discuss.Advance Article, DOI: 10.1039/D4FD00134F (2024)

    2024 doi

  43. [51]

    Merchant, A. et al. Scaling deep learning for materials discovery. Nature 624, 80–85 (2023)

    2023

  44. [52]

    Yang, H. et al. MatterSim: A deep learning atomistic model across elements, tempera- tures and pressures (2024). Preprint at https://arxiv.org/abs/2405.04967

    2024 arXiv

  45. [53]

    Automation in DFT-based computational materials science

    George, J. Automation in DFT-based computational materials science. Trends Chem. 3, 697–699 (2021)

    2021

  46. [54]

    Curtarolo, S. et al. Aflow: An automatic framework for high-throughput materials discovery. Comput. Mater . Sci.58, 218–226 (2012)

    2012

  47. [55]

    Kirklin, S. et al. The Open Quantum Materials Database (OQMD): assessing the accu- racy of DFT formation energies. npj Comput. Mater .1, 15010 (2015)

    2015

  48. [56]

    & Kozinsky, B

    Pizzi, G., Cepellotti, A., Sabatini, R., Marzari, N. & Kozinsky, B. AiiDA: automated interactive infrastructure and database for computational science. Comput. Mater . Sci. 111, 218–230 (2016)

    2016

  49. [57]

    Mathew, K. et al. Atomate: A high-level interface to generate, execute, and analyze computational materials science workflows. Comput. Mater . Sci.139, 140–152 (2017)

    2017

  50. [58]

    Choudhary, K. et al. The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design. npj Comput. Mater .6, 173 (2020)

    2020

  51. [59]

    O., Suh, C., G ´omez-Bombarelli, R., Aguilera-Iparraguirre, J

    Pyzer-Knapp, E. O., Suh, C., G ´omez-Bombarelli, R., Aguilera-Iparraguirre, J. & Aspuru-Guzik, A. What is high-throughput virtual screening? A perspective from organic materials discovery. Annu. Rev. Mater . Res.45, 195–216 (2015)

    2015

  52. [60]

    & Toberer, E

    Gorai, P., Stevanovi ´c, V . & Toberer, E. S. Computationally guided discovery of ther- moelectric materials. Nat. Rev. Mater .2, 17053 (2017)

    2017

  53. [61]

    Finding the needle in the haystack: Materials discovery and design through computational ab initio high-throughput screening

    Hautier, G. Finding the needle in the haystack: Materials discovery and design through computational ab initio high-throughput screening. Comput. Mater . Sci.163, 108–116 (2019)

    2019

  54. [62]

    Janssen, J. et al. pyiron: An integrated development environment for computational materials science. Comput. Mater . Sci.163, 24–36 (2019). 23

    2019

  55. [63]

    DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models

    Zhang, Y .et al. DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models. Comput. Phys. Commun. 253, 107206 (2020)

    2020

  56. [64]

    Gel ˇzinyt˙e, E. et al. wfl Python toolkit for creating machine learning interatomic poten- tials and related atomistic simulation workflows. J. Chem. Phys. 159, 124801 (2023)

    2023

  57. [65]

    & Cheng, J

    Guo, Y .-X., Zhuang, Y .-B., Shi, J. & Cheng, J. ChecMatE: A workflow package to auto- matically generate machine learning potentials and phase diagrams for semiconductor alloys. J. Chem. Phys. 159, 094801 (2023)

    2023

  58. [66]

    Menon, S. et al. From electrons to phase diagrams with classical and machine learning potentials: automated workflows for materials science with pyiron. npj Comput. Mater . 10, 261 (2024)

    2024

  59. [67]

    & Neugebauer, J

    Poul, M., Huber, L. & Neugebauer, J. Automated generation of struc- ture datasets for machine learning potentials and alloys (2024). Preprint at https://www.researchsquare.com/article/rs-4732459/v1

    2024

  60. [68]

    Ganose, A. et al. atomate2 (2024). URL https://github.com/materialsproject/ atomate2

    2024

  61. [69]

    Y ., Stefanoski, S., Kurakevych, O

    Kim, D. Y ., Stefanoski, S., Kurakevych, O. O. & Strobel, T. A. Synthesis of an open- framework allotrope of silicon. Nat. Mater .14, 169–173 (2015)

    2015

  62. [70]

    R., Armstrong, G., Canales, J., Garc ´ıa, R

    Armstrong, A. R., Armstrong, G., Canales, J., Garc ´ıa, R. & Bruce, P. G. Lithium-ion intercalation into TiO2-B nanowires. Adv. Mater .17, 862–865 (2005)

    2005

  63. [71]

    Liang, S. et al. Bronze-phase TiO2 as anode materials in lithium and sodium-ion bat- teries. Adv. Funct. Mater .32, 2201675 (2022)

    2022

  64. [72]

    P., Burke, K

    Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996)

    1996

  65. [73]

    & Perdew, J

    Sun, J., Ruzsinszky, A. & Perdew, J. P. Strongly constrained and appropriately normed semilocal density functional. Phys. Rev. Lett. 115, 036402 (2015)

    2015

  66. [74]

    &´Angy´an, J

    Demuth, T., Jeanvoine, Y ., Hafner, J. &´Angy´an, J. G. Polymorphism in silica studied in the local density and generalized-gradient approximations. J. Phys.: Condens. Matter 11, 3833–3874 (1999)

    1999

  67. [75]

    C., Rohrer, J., Albe, K

    Erhard, L. C., Rohrer, J., Albe, K. & Deringer, V . L. Modelling atomic and nanoscale structure in the silicon–oxygen system through active machine learning. Nat. Commun. 15, 1927 (2024). 24

    2024

  68. [76]

    Markland, T. E. & Ceriotti, M. Nuclear quantum effects enter the mainstream. Nat. Rev. Chem. 2, 0109 (2018)

    2018

  69. [77]

    The radial distribution functions of water and ice from 220 to 673 K and at pressures up to 400 MPa

    Soper, A. The radial distribution functions of water and ice from 220 to 673 K and at pressures up to 400 MPa. Chem. Phys. 258, 121–137 (2000)

    2000

  70. [78]

    B., Benmore, C

    Skinner, L. B., Benmore, C. J., Neuefeind, J. C. & Parise, J. B. The structure of water around the compressibility minimum. J. Chem. Phys. 141, 214507 (2014)

    2014

  71. [79]

    P., Hutter, J., Kuo, I.-F

    Todorova, T., Seitsonen, A. P., Hutter, J., Kuo, I.-F. W. & Mundy, C. J. Molecular dynamics simulation of liquid water: Hybrid density functionals. J. Phys. Chem. B 110, 3685–3691 (2006)

    2006

  72. [80]

    G., Engel, E

    Monserrat, B., Brandenburg, J. G., Engel, E. A. & Cheng, B. Liquid water contains the building blocks of diverse ice phases. Nat. Commun. 11, 5757 (2020)

    2020

  73. [81]

    Thompson, A. P. et al. LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. Comp. Phys. Comm. 271, 108171 (2022)

    2022

  74. [82]

    Gardner, J. L. A. graph-pes: train and use graph-based Machine-Learned models of Potential Energy Surfaces. https://github.com/jla-gardner/graph-pes (2024). Version 0.0.10

    2024

  75. [83]

    H., Sciortino, F., Grande, T., Stanley, H

    Poole, P. H., Sciortino, F., Grande, T., Stanley, H. E. & Angell, C. A. Effect of hydrogen bonds on the thermodynamic behavior of liquid water.Phys. Rev. Lett. 73, 1632 (1994)

    1994

  76. [84]

    & Skinner, J

    Kumar, R., Schmidt, J. & Skinner, J. Hydrogen bonding definitions and dynamics in liquid water. J. Chem. Phys. 126, 204107 (2007)

    2007

  77. [85]

    Salzbrenner, P. T. et al. Developments and further applications of ephemeral data de- rived potentials. J. Chem. Phys. 159, 144801 (2023)

    2023

  78. [86]

    Gardner, J. L. A., Baker, K. T. & Deringer, V . L. Synthetic pre-training for neural- network interatomic potentials. Mach. Learn.: Sci. Technol. 5, 015003 (2024)

    2024

  79. [87]

    & Yamada, N

    Wuttig, M. & Yamada, N. Phase-change materials for rewriteable data storage. Nat. Mater .6, 824–832 (2007)

    2007

  80. [88]

    Selmo, S. et al. Low power phase change memory switching of ultra-thin In 3Sb1Te2 nanowires. Appl. Phys. Lett. 109, 213103 (2016)

    2016

  81. [89]

    Zhang, W., Mazzarello, R., Wuttig, M. & Ma, E. Designing crystallization in phase- change materials for universal memory and neuro-inspired computing. Nat. Rev. Mater . 4, 150–168 (2019). 25

    2019

  82. [90]

    Y .et al

    Raty, J. Y .et al. Aging mechanisms in amorphous phase-change materials. Nat. Com- mun. 6, 7467 (2015)

    2015

  83. [91]

    H., K ¨uhne, T

    Los, J. H., K ¨uhne, T. D., Gabardi, S. & Bernasconi, M. First-principles study of the amorphous In3SbTe2 phase change compound. Phys. Rev. B 88, 174203 (2013)

    2013

  84. [92]

    P., Kondor, R

    Bart ´ok, A. P., Kondor, R. & Cs´anyi, G. On representing chemical environments. Phys. Rev. B 87, 184115 (2013)

    2013

  85. [93]

    Xu, Y . et al. Unraveling crystallization mechanisms and electronic structure of phase- change materials by large-scale ab initio simulations. Adv. Mater .34, 2109139 (2022)

    2022

  86. [94]

    L., Dronskowski, R

    Deringer, V . L., Dronskowski, R. & Wuttig, M. Microscopic complexity in phase- change materials and its role for applications.Adv. Funct. Mater .25, 6343–6359 (2015)

    2015

  87. [95]

    Schr ¨oder, T. et al. Disorder and transport properties of In 3SbTe2 – an X-ray, neutron and electron diffraction study. Z. Anorg. Allg. Chem. 639, 2536–2541 (2013)

    2013

  88. [96]

    L., Caro, M

    Deringer, V . L., Caro, M. A. & Cs ´anyi, G. A general-purpose machine-learning force field for bulk and nanostructured phosphorus. Nat. Commun. 11, 5461 (2020)

    2020

  89. [97]

    D., Gardner, J

    Morrow, J. D., Gardner, J. L. A. & Deringer, V . L. How to validate machine-learned interatomic potentials. J. Chem. Phys. 158, 121501 (2023)

    2023

  90. [98]

    F., Zhou, Y

    Thomas du Toit, D. F., Zhou, Y . & Deringer, V . L. Hyperparameter optimization for atomic cluster expansion potentials.J. Chem. Theory Comput. 20, 10103–10113 (2024)

    2024

  91. [99]

    Witt, W. C. et al. ACEpotentials.jl: A Julia implementation of the atomic cluster ex- pansion. J. Chem. Phys. 159, 164101 (2023)

    2023

  92. [100]

    Rosen, A. S. et al. Jobflow: Computational Workflows Made Simple. J. Open Source Softw. 9, 5995 (2024)

    2024

  93. [101]

    Jain, A. et al. FireWorks: a dynamic workflow system designed for high-throughput applications. Concurr . Comput.27, 5037 (2015)

    2015

  94. [102]

    Ong, S. P. et al. Python Materials Genomics (pymatgen): A robust, open-source python library for materials analysis. Comput. Mater . Sci.68, 314–319 (2013)

    2013

  95. [103]

    Hjorth Larsen, A. et al. The Atomic Simulation Environment—a Python library for working with atoms. J. Phys.: Condens. Matter 29, 273002 (2017)

    2017

  96. [104]

    Bl ¨ochl, P. E. Projector augmented-wave method.Phys. Rev. B 50, 17953–17979 (1994)

    1994

  97. [105]

    & Joubert, D

    Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented- wave method. Phys. Rev. B 59, 1758–1775 (1999). 26

    1999

  98. [106]

    & Furthm ¨uller, J

    Kresse, G. & Furthm ¨uller, J. Efficient iterative schemes for ab initio total-energy cal- culations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996)

    1996

  99. [107]

    Perdew, J. P. et al. Restoring the Density-Gradient Expansion for Exchange in Solids and Surfaces. Phys. Rev. Lett. 100, 136406 (2008)

    2008

  100. [108]

    J., Pickard, C

    Lyle, M. J., Pickard, C. J. & Needs, R. J. Prediction of 10-fold coordinated TiO 2 and SiO2 structures at multimegabar pressures. Proc. Natl. Acad. Sci. USA 112, 6898–6901 (2015)

    2015

  101. [109]

    Zhang, L., Wang, H., Car, R. & E, W. Phase diagram of a deep potential water model. Phys. Rev. Lett. 126, 236001 (2021)

    2021

  102. [110]

    Visualization and analysis of atomistic simulation data with OVITO–the Open Visualization Tool

    Stukowski, A. Visualization and analysis of atomistic simulation data with OVITO–the Open Visualization Tool. Model. Simul. Mater . Sci. Eng.18, 015012 (2009). 27

    2009

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.