OK, so, after reading some discussion here I think everyone should calm down and think about this proposal with cool head. I've jumped in and was pretty excited about it at first, but now I'm becoming more and more skeptical. I think we should consider the following reasons before actually committing to the new site:

  1. I don't see a single question in proposed list which is not appropriate for Cross Validated. So, there exists a site where all these questions could be asked and where all the same people (plus more) are present. So it doesn't give any improvements on coverage of topics or users...

  2. Cross Validated is not hugely popular. It is just OK. And machine learning is 4th most popular tag there. That means we are going to split out a big chunk of CV, which will make two semi-dead pages out of one OK page. This is considering people will actually move to ML, I feel like most will just keep visiting CV and ML will have an extremely low traffic. I'm not even starting on confusion about where to actually post the ML-related questions (given the already popular category on CV). What will happen is that most of ML guys will be registered on both and newcommers will be pretty confused.

I'm writing all this simply because I think whatever is done here should be done for the benefit of the community. The reasons for opening a new page would be if there is a big area of questions which are not covered in already existing SE page or if there is next to none overlap of this area with already existing site. Neither of these is true in this case. "We want to get apart from those linear regression guys" is not a valid reason.

Don't get me wrong, if it does go through I'm going to participate actively for sure, I just don't think it is going to improve anything. I'm passionate about machine learning and do understand when people are excited about it, but I'm also passionate about SE and in my opinion emotions are not what should be the motivation.

All in all, please think in terms of benefit to community first, and then vote. I'm happy to discuss further if people have different opinion.

Proposal: Machine Learning

  • I disagree that this would divide CV into two semi-dead pages. ML will have massive crossovers with not just CV, but SO, EE, and robotics. And its not like users have to choose one. Many users use multiple sites, and like to keep them separated by relevance so they can get the information they want. – travisbartley Jun 6 '13 at 4:43
  • @trav1s So, some questions (or many) will migrate from one page to another, that's all. This will make ALL of the aforementioned sites less active, not more active. – sashkello Jun 9 '13 at 3:54
  • The motivation is primarily to make highly relevant sites in which user communities can be built, not to make one high-activity site. Under your logic most SE sites would have never splintered from SO. They did so to reduce irrelevant noise to the user. In doing so, SO was made less active, but with no net loss in activity to SE. In fact, the overall SE activity would clearly increase if each site is more relevant to its respective user base. – travisbartley Jun 10 '13 at 6:07
  • 1
    @trav1s Not exactly. I don't see any SE site which was built for such a cause. It would then split into Java, C++, etc., but the thing is that there is a lot of common knowledge in between these things and that's why they are in one place. Same with machine learning and stats or CS. They simply can't be separated, because in this case it is literally a subfield of one or another. – sashkello Jun 10 '13 at 8:24
  • 1
    In addition to this, how many sites are you really ready to be active at? I don't think anyone would seriously be active at more than 3. If they keep splitting it will shift focus and confuse people. I wouldn't want to jump from CS to Mathematics to CV to ML and think where my question belongs (gosh, if there is some programming thrown in there, should I go to SO then...). As I said, there should be one perfect place to ask each type of questions, and it should be clear. It is not clear where to ask Random Forest classification question because it belongs both to stats and ML. – sashkello Jun 10 '13 at 8:28
  • 1
    I still haven't seen a single question which one can't ask already and which will be welcome on ML. None. The sites are already there, the people are already there. And returning to your first point, also I don't know a single site which did split from SO on which questions fully belong to SO. – sashkello Jun 10 '13 at 8:30
  • You might be right that ML is too similar to CV... In that case, we should change the name from ML to AI to achieve greater separation in content. AI uses statistical methods, but is not a subcategory of statistics. – travisbartley Jun 11 '13 at 5:19
  • @trav1s Totally agree, I'd completely support Robotics/AI/etc proposal, but I think there were a few of those already... – sashkello Jun 11 '13 at 5:21
  • Yes, and there will be more proposals after this one fails to reach the critical threshold. – travisbartley Jun 11 '13 at 5:23
  • Is it even possible to change the name and definition from ML to AI after reaching the commitment phase? If not, we should trash this proposal and start a new one. – travisbartley Jun 11 '13 at 5:41
  • Well, as it stands, it will go into live stage and will be inactive for a while and be closed again, as it was not so long before :) – sashkello Jun 11 '13 at 5:48
  • Note: New Artificial Intelligence site is beta now, so this can be discussed whether machine learning should be on-topic there. – kenorb Aug 10 '16 at 11:09

There is already an ML tag in CV (205 followers, 1164 questions; total CV users ~17k) and one in CS beta (it's ranked 25 of 432, 30 followers, 84 questions; total CS users ~6k). Presumably CS will grow once it goes public and so will followers of the ML tag there. I agree with those here who say that splitting out ML will dilute the exposure. If I had a mathematically slanted ML question, I'd go to the CV board. If it was CS slanted, I'd go to the CS board. Regardless of how things are classified, there are already places to get those questions answered. The assertion that CV is "really ... a machine learning board" only supports the argument that creating a separate ML board is not in the best interest of the community because people already know where to go for answers. So,I think the best answer to the question "Why do we need it?" is that we don't.

  • 1
    Don't forget about programming & algorithms related ML questions on SO. – sashkello May 24 '13 at 13:57
  • Good point. That adds another 1700 followers and 2719 questions. – cb4 May 25 '13 at 16:57

I think that there is a good chance that an ML site would incorporate theoretical questions from MSE like this one or this one, and implementation questions like this one from CV.

I asked or answered all of the linked questions above, and it would be nice to see them answered in their own area, or given attention as such (I believe I got Tumbleweed for one of them). Mathematicians, for one, may be less interested in some of the shaky ground on which adaptive filters like LMS or other learning algorithms are based.

  • 2
    There are LOTS of unanswered questions on Math and CV, that doesn't mean separating each of those topics will increase the chance of getting an answer, IMO, it only decreases that chance (in addition, one of presented questions is answered/accepted same day which goes counter your point). There are not only mathematicians on Math and not only statisticians on CV and evidently a lot of ML experts are among them (these are not outsiders who join this proposal). (I'm actually a physicist) There are many topics some people don't care about which doesn't restrict others from answering. – sashkello Jun 19 '13 at 2:44
  • I would agree with you up to one point--how do you gauge how much ML interest there is (i.e., what's the ML specialist join rate to each site), and how do you determine the expected increase (or lack there of) of ML specialists in the event an ML site was created. Personally I feel like CV and ML don't attract ML people enough, but I don't have evidence for it. – bright-star Jun 19 '13 at 6:49

The real answer to your question is that CV is really (so I would argue) a machine learning board. It seems to be pretty consistently concerned with statistical inference and, more particularly, the methodology of practical statistical inference, which are the primary concern of ML scientists. It is not usually primarily concerned with things like asymptotics, convergence, guarantees, and other things that dyed-in-the-wool statisticians are usually concerned with.

But here's the problem. It is a good ML board, but it is not billed as a ML board. People don't think of it as an ML board, and they don't advertise it as an ML board.

The reason this is a problem is because, especially for lay people, it is not clear where to go to talk about ML. MetaOptimize? CV? Some Google+ group? Which place has the best community? Which is most likely to answer your questions? One of your questions is whether there's enough room in this town for all of us. That there are already so many medium-traffic communities indicates that, yes, there is. The question for me is: where have these communities failed, and how can we build a better one? Because, IMHO, in one way or another, they have indeed all failed.

Of the existing communities, I think CV is the best, but it is not clear that people would know to go there, and it is not clear that the community would want to be called an "ML" community, since statisticians -- which is what they would say they are, I think -- think of ML as a faddy renaming for vanilla inference.

  • 1
    I agree with most of things you said. In such terms it would be much more beneficial to alter the CV statement and try to attract ML people there, rather than create a new page. Because now we'll have two pages where you can ask ML questions, the only difference is the name... – sashkello Apr 18 '13 at 2:36
  • Oh I totally agree. But CV is not going to rename themselves as an ML community. So the question we should ask is what we want to do, since this is virtually certain? – apc Apr 18 '13 at 6:58
  • 3
    @apc But we already did that! Like at the beginning of public beta it was justified that this is stats+ML site, and since then it is explicitly stated in CV's FAQ, about page and via moderating policy. Even more, this is why we fought so hard to make it called ambiguous "CrossValidated" rather than "Stats SE". – mbq Apr 27 '13 at 17:56
  • 1
    That's the opposite of what I am saying. I'm saying a stats+ML community is a barrier. We just want an ML community. – apc Apr 29 '13 at 5:49
  • 2
    We are returning to the initial question: WHY??? Where does classification using Random Forest belongs to? It is classification question, so CV is perfect but it is also machine learning... Regression would go to CV though, wouldn't it? The boundary is non-existent. Almost any question I can imagine is a cross-over. – sashkello Apr 29 '13 at 23:28
  • I don't understand how you are confused about my position on this issue. I think I've been pretty clear about this. ML is a type of statistical inference, so of course it is acceptable to propose under CV. Please help me to understand where you are having trouble. – apc Apr 30 '13 at 5:44
  • 2
    You say "We just want an ML community" and "ML is a type of statistical inference" at the same time. Why would we want a separate community while it is included in CV and most of questions are hardly classifiable as one or another. – sashkello Apr 30 '13 at 23:32
  • 2
    @apc Note that there was a site called “Machine learning” not so long ago and that it didn't work out. Other users seem to think that we should keep trying no matter what and eventually it could work but it's difficult to think that using this or that name would be enough to bring it off the ground. – Gala May 7 '13 at 19:17
  • 1
    Also, I basically share sashkello's concerns. It would conceivably be worth trying if there was noting to loose by doing it but two sites means less exposure to each question, a smaller reserve of past questions and answers to find relevant material, etc. – Gala May 7 '13 at 19:20
  • 1
    That CV is not advertised as an ML board is not relevant. Between the CV and CS boards, ML questions can get answered. – cb4 May 23 '13 at 21:54

You must log in to answer this question.

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