I would encode DFAs (Deterministic Finite State automata) as output (or input) of a neural network for a supervised learning; it is well-known [1] that efficacy of neural network training strongly depends on adopted encoding.

How can I encode DFAs for a neural network? Are there any literature works?

I've already found some algorithms being able to extract a DFA from a recurrent neural network, but nothing about DFAs either as input or output of ANN.

[1]: Neural network encoding approach comparison: an empirical study; Investigating the Effect of Data Representation On Neural Network and Regression; Data Mining and Neural Networks: The Impact of Data Representation.

  • 4
    $\begingroup$ Use any encoding. What's the problem/question here? $\endgroup$ – Raphael Jul 20 '16 at 11:38
  • $\begingroup$ We expect references to fulfill the minimal scholarly requirements and be as robust over time as possible. Please take some time to improve your post in this regard. We have collected some advice here. $\endgroup$ – Raphael Jul 21 '16 at 9:49
  • $\begingroup$ Just in time ;) $\endgroup$ – Gabrer Jul 21 '16 at 9:53
  • $\begingroup$ I encourage you to read the link Raphael suggested. We'd like references to be robust over time, which means more than just a link: I suggest listing title, authors, and where published. (Title isn't always enough to uniquely find a paper, unfortunately.) $\endgroup$ – D.W. Jul 21 '16 at 18:04
  • $\begingroup$ As it stands I don't think the problem is answerable. With only the information you've given us, Raphael has exactly the right answer: we have no basis for suggesting any one encoding over another. If you want help identifying encodings that might be more suitable, you'll need to also tell us the task: what are you trying to learn with machine learning? Yes, the efficiency of machine learning can be dependent on how you encode inputs and what you select as features, but usually features need to be designed with the task in mind. $\endgroup$ – D.W. Jul 21 '16 at 18:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.