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.

Artificial intelligence

Deep neural reasoning

Nature volume 538, pages 467–468 (2016)Cite this article

Subjects

The human brain can solve highly abstract reasoning problems using a neural network that is entirely physical. The underlying mechanisms are only partially understood, but an artificial network provides valuable insight. See Article p.471

A classic example of logical reasoning is the syllogism, “All men are mortal. Socrates is a man. Therefore, Socrates is mortal.” According to both ancient and modern views1, reasoning amounts to a rule-based mental manipulation of symbols — in this example, the words 'All', 'men', and so on. But human brains are made of neurons that operate by exchanging jittery electrical pulses, rather than word-like symbols. This difference encapsulates a notorious scientific and philosophical enigma, sometimes referred to as the neural–symbolic integration problem2, which remains unsolved. On page 471, Graves et al.3 use the machine-learning methods of 'deep learning' to impart some crucial symbolic-reasoning mechanisms to an artificial neural system. Their system can solve complex tasks by learning symbolic-reasoning rules from examples, an achievement that has potential implications for the neural–symbolic integration problem.

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.

Notes

  1. See all news & views

References

  1. Newell, A. Cogn. Sci. 4, 135–183 (1980).

    Article  Google Scholar 

  2. Hammer, B. & Hitzler, P. (eds) Perspectives of Neural–Symbolic Integration http://doi.org/fsrb8m (Springer, 2007).

    Book  Google Scholar 

  3. Graves, A. et al. Nature 538, 471–476 (2016).

    ADS  Article  Google Scholar 

  4. Turing, A. M. J. Math. 58, 345–363 (1936).

    Article  Google Scholar 

  5. LeCun, Y., Bengio, Y. & Hinton, G. Nature 521, 436–444 (2015).

    ADS  CAS  Article  Google Scholar 

  6. Szegedy, C. et al. Proc. IEEE Conf. Computer Vision Pattern Recognition http://dx.doi.org/10.1109/CVPR.2015.7298594 (2015).

  7. Bahdanau, D., Cho, K. & Bengio, Y. Int. Conf. Learning Representations Preprint at http://arxiv.org/abs/1409.0473 (2014).

  8. De Raedt, L. & Kimmig, A. Machine Learn. 100, 5–47 (2015).

    MathSciNet  Article  Google Scholar 

  9. Graves, A., Wayne, G. & Danihelka, I. Preprint at http://arxiv.org/abs/1410.5401 (2014).

Download references

Author information

Authors and Affiliations

  1. Herbert Jaeger is at Jacobs University Bremen, 28759 Bremen, Germany.,

    Herbert Jaeger

Authors
  1. Herbert Jaeger
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Herbert Jaeger.

Related audio

Related links

Related links

Related links in Nature Research

Computer science: Nanoscale connections for brain-like circuits

Artificial intelligence: Learning to see and act

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Jaeger, H. Deep neural reasoning. Nature 538, 467–468 (2016). https://doi.org/10.1038/nature19477

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nature19477

Further reading

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