Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

COMPUTATIONAL PHYSICS

To V or not to V

Nature Computational Science volume 1, pages 644–645 (2021)Cite this article

Subjects

Finding a parameter that can accurately identify the order–disorder phase transition, especially for complex physical systems with high-dimensional configurational space, is a challenging task. Recent work proposes a machine learning approach to effectively tackle this challenge.

Phases of matter and the transitions between them are the central concepts in solid-state physics. Whether it is a transition of a metal into a superconducting state, a transition from a ferromagnetic to a paramagnetic state, or the transition between an ordered and a disordered lattice, the identification of such transitions typically relies on finding an order parameter. The order parameter is a computable measure that distinguishes two different phases, having a value of zero in one phase and a non-zero value in another. Conventionally, order parameters require ad hoc assumptions, especially with experts’ insight about the physical system. Writing in Nature Computational Science, Yin and colleagues introduce a machine learning-based method to gain such insights from the data instead, with the goal of computing generic order parameters for the order–disorder transition in high-entropy alloys (HEAs)1.

Your institute does not have access to this article

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

Buy

All prices are NET prices.

Fig. 1: The VAE-based order parameter.

References

  1. Yin, J., Pei, Z. & Gao, M. C. Nat. Comput. Sci. https://doi.org/10.1038/s43588-021-00139-3 (2021).

  2. George, E. P., Raabe, D. & Ritchie, R. O. Nat. Rev. Mater. 4, 515–534 (2019).

    Article  Google Scholar 

  3. Kingma, D. P. & Welling, M. Auto-encoding variational Bayes. In 2nd ICLR Conference Track Proceedings (ILCR, 2014).

  4. Wetzel, S. J. Phys. Rev. E 96, 022140 (2017).

    Article  Google Scholar 

  5. Schmitt, M. & Lenarčič, Z. Preprint at https://arxiv.org/abs/2102.11328 (2021).

  6. van der Maaten, L. J. P. & Hinton, G. E. J. Mach. Learn. Res. 9, 2579–2605 (2008).

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark

    Evert van Nieuwenburg

Authors
  1. Evert van Nieuwenburg
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Evert van Nieuwenburg.

Ethics declarations

Competing interests

The author declares no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

van Nieuwenburg, E. To V or not to V. Nat Comput Sci 1, 644–645 (2021). https://doi.org/10.1038/s43588-021-00143-7

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s43588-021-00143-7

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing