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.

  • NEWS AND VIEWS

AI learns to write sorting software on its own

Deep reinforcement learning has been used to improve computer code by treating the task as a game — with no special knowledge needed on the part of the player. The result has already worked its way into countless programs.
  1. Armando Solar-Lezama

    1. Armando Solar-Lezama is in the Computer Science & Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.

    View author publications

    You can also search for this author in PubMed  Google Scholar

For decades, the computing industry relied on Moore’s law: as transistors became ever smaller, the number that could be crammed onto a computer chip seemed to double every two years, enabling a similar leap in computing power. But Moore’s law has a natural limit, so software optimization has become just as crucial as miniaturization. Writing in Nature, Mankowitz et al.1 reveal a key role for deep learning in this process, by showing that code generated by artificial intelligence (AI) can improve the efficiency with which the C++ programming language sorts items in a list. Although seemingly mundane, this task is needed in computer programs the world over, and the AI version is now baked into a widely used implementation of the C++ library. Perhaps even more remarkably, the AI system can improve the code without any previous knowledge of the problem itself.

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Nature 618, 240-241 (2023)

doi: https://doi.org/10.1038/d41586-023-01812-5

References

  1. Mankowitz, D. J. et al. Nature 618, 257–263 (2023).

    Article  Google Scholar 

  2. Bacon, D. F., Graham, S. L. & Sharp, O. J. ACM Comput. Surv. 26, 345–420 (1994).

    Article  Google Scholar 

  3. Balaprakash, P. et al. Proc. IEEE 106, 2068–2083 (2018).

    Article  Google Scholar 

  4. Vuduc, R. W. in Encyclopedia of Parallel Computing (ed. Padua, D.) 102–105 (Springer, 2011).

    Google Scholar 

  5. Frigo, M. ACM SIGPLAN Not. 34, 169–180 (1999).

    Article  Google Scholar 

  6. Whaley, R. C., Petitet, A. & Dongarra, J. J. Parallel Comput. 27, 3–35 (2001).

    Article  Google Scholar 

  7. Püschel, M. et al. Proc. IEEE 93, 232–275 (2005).

    Article  Google Scholar 

  8. Franchetti, F. & Püschel, M. Proc. Int. Parallel Distributed Process. Symp. (IEEE, 2003).

    Google Scholar 

  9. Alur, R. et al. in Formal Methods in Computer-Aided Design 2013 1–8 (IEEE, 2013).

    Google Scholar 

  10. Barthe, G., Crespo, J. M., Gulwani, S., Kunz, C. & Marron, M. ACM SIGPLAN Not. 48, 123–134 (2013).

    Article  Google Scholar 

  11. Schkufza, E., Sharma, R. & Aiken, A. in Proc. 18th Int. Conf. Archit. Support for Program. Lang. Oper. Syst. 305–316 (IEEE, 2013).

    Google Scholar 

  12. Devlin, J. et al. in Proc. 34th Int. Conf. Mach. Learn. Vol. 70 (eds Precup, D. & Teh, Y. W.) 990–998 (JMLR, 2017).

    Google Scholar 

  13. Li, Y. et al. Science 378, 1092–1097 (2022).

    Article  PubMed  Google Scholar 

  14. Ellis, K. et al. in Proc. 33rd Neural Inf. Process. Syst. (eds Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F. & Fox, E. B.) 9169–9178 (Curran Associates, 2019).

    Google Scholar 

Download references

Competing Interests

A.S.-L. consults very sporadically with Google X (about 10 hours in the past year). Its parent company, Alphabet, is the same as that of DeepMind.

Subjects

Latest on:

Nature Careers

Jobs

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

Search

Advanced search

Quick links