I would like learn about use neural networks in cryptography. Is there any research literature on this subject? What should I do to learn about the subject?

I am taking short course in neural networks and for the course project I would like to do something in neural cryptography.

  • 3
    $\begingroup$ Have you done a background/lit. search? If so, would you mind sharing some of the results you've come across. For me in particular and probably others here, neural crypto is something I've never heard of. $\endgroup$ – mikeazo May 10 '13 at 14:02
  • $\begingroup$ @mikeazo en.wikipedia.org/wiki/Neural_cryptography it apparently exists, though I've never heard of it either. Sounds like it would be more appropriate to cryptanalysis than the actual process of using cryptography. $\endgroup$ – Thomas May 10 '13 at 14:08
  • $\begingroup$ Yep, it is something new, and not so much research has been done in this domain. I even failed to create new tag "neural" and "neural network", because I have no reputation, I am new here. So, a few links: scs-europe.net/conf/ecms2012/ecms2012%20accepted%20papers/… ics.org.ru/doc?pdf=857&dir=e ki.tu-berlin.de/fileadmin/fg135/publikationen/… en.wikipedia.org/wiki/Neural_cryptography $\endgroup$ – Jager May 10 '13 at 14:18
  • 2
    $\begingroup$ This site is not a good place for general, open-ended discussions ("I would like you to express your opinions about [broad subject]"). Instead, it is for concrete questions with a correct answer. Please read the FAQ before asking questions. $\endgroup$ – D.W. May 10 '13 at 19:07
  • $\begingroup$ @D.W., is right. The only reason I wrote an answer is because it wouldn't fit in a comment. $\endgroup$ – mikeazo May 10 '13 at 19:22

It's got a wikipedia page so it must be "serious" :)

From my very quick look it seems like a field that isn't too new (90's). The paper Analysis of Neural Cryptography has Adi Shamir's name on it (the "S" in RSA and the Shamir from Shamir secret sharing), so there has at least been a very reputable cryptographer interested in the idea at one point. Searches on IACR's ePrint archive turn up very little (one hit with "neural" in the Anywhere field).

So, the field seems to be not very well explored and has not generated broad interest. That said, if you are taking the class, have to do some kind of project, and are interested in cryptography, why not find something related? If the project doesn't have to be too earth shattering, I'd try attacking some "classical" ciphers or even see if you can classify classical ciphers (Caesar, vigenere, etc) based on the ciphertext only. Another interesting area would be attacking something like Enigma with neural networks.

Just my 2 cents.


Not the answer you're looking for? Browse other questions tagged or ask your own question.