Machine learning glasses

Artificial neural networks now allow the dynamics of supercooled liquids to be predicted from their structure alone in an unprecedented way, thus providing a powerful new tool to study the physics of the glass transition.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.


  1. 1.

    Bapst, V. et al. Nat. Phys. (2020).

  2. 2.

    Widmer-Cooper, A., Harrowell, P. & Fynewever, H. Phys. Rev. Lett. 93, 135701 (2004).

    ADS  Article  Google Scholar 

  3. 3.

    Sussman, D. M., Schoenholz, S. S., Cubuk, E. D. & Liu, A. J. Proc. Natl Acad. Sci. USA 114, 10601–10605 (2017).

    ADS  MathSciNet  Article  Google Scholar 

  4. 4.

    Ninarello, A., Berthier, L. & Coslovich, D. Phys. Rev. X 7, 021039 (2017).

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Giulio Biroli.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Biroli, G. Machine learning glasses. Nat. Phys. 16, 373–374 (2020).

Download citation


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing