14
$\begingroup$

The question "What is sigma function in the YOLO object detector?" has been closed as off-topic with the usual explanation:

This question appears to be off-topic because EITHER it is not about statistics, machine learning, data analysis, data mining, or data visualization, OR it focuses on programming, debugging, or performing routine operations within a statistical computing platform. If the latter, you could try the support links we maintain.

However, I think it does not really fit this description:

  • It is arguably about machine learning.
  • It is not about debugging neither performing routine operations within a statistical platform.

It clearly asks for explanation of mathematical notation used in one particular paper. It has even got two answers (one being accepted) before being closed. For the OP, it makes no difference whether it stays closed or not, but for any future visitors, it sends a message that "we don't want such questions here". In my opinion, it is not the case—except for using a screenshot instead of writing down the equations, it is an on-topic and answerable question.

Before I vote to reopen, I would like to know if I may have overlooked any other reason for off-topicness of this particular question.

$\endgroup$

1 Answer 1

Reset to default
17
$\begingroup$

Questions about notation have been traditionally considered within our scope, provided the notation is ultimately about statistics / machine learning. In addition, the question of why the exponential is used seems to be a meaningful ML question. I don't see that it needs to be closed. I just reopened it.

$\endgroup$
5
  • 2
    $\begingroup$ I agree with this action. But the post doesn't strike me as being a very good question, a quite different issue. It seems that the original authors didn't define their notation but my guess is that they would regard what they used as standard in their field. More directly put, would we be happy with posts asking for definitions of $\pi$ or $\exp()$ because authors didn't explain them? Are we here to teach people the little bits of mathematics they should be finding out any way, e.g. by reading simpler papers, textbooks, Wikipedia? $\endgroup$
    – Nick Cox
    Apr 29, 2019 at 15:18
  • 1
    $\begingroup$ (I stopped formal study of mathematics at 17, so the previous is not "Let them all learn mathematics" from someone with a mathematics or statistics degree.) $\endgroup$
    – Nick Cox
    Apr 29, 2019 at 15:19
  • 10
    $\begingroup$ @NickCox, under the assumption that that happens, we would have $1$ question asking what $\pi$ is & $1$ question asking what $\exp$ is (all subsequent Qs should end up marked as duplicates). Personally, I would be fine with that. We don't have a lower bound on questions that are 'too trivial' & thus off topic for that reason. (OTOH, downvoting is fine for Qs that you believe 'lack research effort'). $\endgroup$
    – gung - Reinstate Monica
    Apr 29, 2019 at 15:22
  • $\begingroup$ I think that's a fair summary. $\endgroup$
    – Nick Cox
    Apr 29, 2019 at 15:30
  • $\begingroup$ Though if a $\pi$ question pointed at something which actually said, for example, "the prior density $\pi_0(\theta)$", it might not be the same question as a question asking about the $\pi$ in "the density $\dfrac{1}{\sqrt{2 \pi} \sigma}\exp\left(-\dfrac{(x-\mu)^2}{2\sigma^2}\right)$" $\endgroup$
    – Henry
    May 4, 2019 at 1:16

You must log in to answer this question.

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