I was curious about this exact question: What are the advantages of ReLU over sigmoid function in deep neural networks?

Reading through there seems to be some confusion, mainly from user Alex R., who doesn't seem happy with many of the answers there. Although they're not happy, they've not left an answer of their own, which leaves me unsure of what I can really trust from the post. The user seems to have a fairly decent reputation, so I'd be uncomfortable with dismissing their views on the thread.

It seems that making another post "because I don't trust the last one" would be a bit off, so I'm unsure what the best thing in this scenario is.

  • 17
    $\begingroup$ You can add a bounty to the question and explain that you would like an authoritative answer with citations, perhaps explaining what, exactly, you'd like to know more about in the bounty message. $\endgroup$ – Sycorax says Reinstate Monica Jan 2 at 16:52
  • $\begingroup$ Notice though that @AlexR says one of the answers is the correct one in his opinion $\endgroup$ – Firebug Jan 13 at 22:00
  • $\begingroup$ @Firebug given the disparity between what he's stated and where the votes are placed it seems a bit confusing to me still... are these votes because people understand the problem and have agreed with the top-level post? Or are they beginners like myself who are just saying thanks (and leaving with the wrong info as a result). $\endgroup$ – baxx Jan 13 at 22:24
  • $\begingroup$ Do you trust more AlexR or the rest of the answers? You must be able to judge this yourself. $\endgroup$ – Firebug Jan 14 at 15:45
  • $\begingroup$ Also, notice that if you created a "new" question, AlexR already said he thinks that one answer is correct. So adding the "new" question would be pointless anyways. $\endgroup$ – Firebug Jan 14 at 15:46

You must log in to answer this question.

Browse other questions tagged .