It seems as if the tag could always be replaced by . Should we remove the tag and add a notice that the should be used instead?

  • $\begingroup$ (there is the same issue on the statistics Stack Exchange) $\endgroup$ Commented May 1, 2017 at 2:14

1 Answer 1


Although they happen to overlap a lot, I think they're still distinct. I would consider something like the Perceptron a neural network but not an example of a deep learning algorithm. Unless the tags are obviously redundant I wouldn't merge them.

  • $\begingroup$ Deep learning is certainly the most important sub-category of neural networks (or is there any deep-learning question which is not also a neural network question?). But I don't think there are more than a hand full of questions which are about neural networks but not deep learning. And many questions which are tagged only with neural network should be tagged with deep learning. Even questions with "deep" in the title often don't have the "deep-learning" tag. $\endgroup$ Commented Apr 25, 2017 at 14:23
  • $\begingroup$ I'm not sure what to do in this situation. I agree they're nearly synonymous but don't know if we should forcibly redirect neural-networks to deep-learning. There are thousands of similar instances, like, shouldn't all javaee questions be tagged java or something? I don't know. $\endgroup$
    – Sean Owen
    Commented Apr 25, 2017 at 14:54
  • $\begingroup$ I think this depends on the size of the set of "dividing" questions and it might change by time. By now, it seems to me that having multiple tags might rather harm because the community is small and people who follow one tag might not follow the other. Probably the tag description of one should mention the other so that people who search can also notice that their answer might have another tag? (But at the end, I guess it doesn't matter too much) $\endgroup$ Commented Apr 25, 2017 at 16:04
  • 1
    $\begingroup$ It probably depends a little on whether people using 'Deep Learning' as a tag are usually asking questions about issues that specifically relate to deep learning - such as around convolutional networks or autoencoders, rather than say explanations of particular perceptrons or general advice on building neural nets. $\endgroup$ Commented May 5, 2017 at 4:14
  • $\begingroup$ Sean gave an example of neural without deep, there is also deep without neural. $\endgroup$ Commented Jul 25, 2017 at 19:06

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .