bridging the gap between machine learning and statistics

sunwar2Today in Warwick, I had a very nice discussion with Michael Betancourt on many statistical and computational issues but at one point in the conversation we came upon the trouble of bridging the gap between the machine learning and statistics communities. While a conference like AISTATS is certainly contributing to this, it does not reach the main bulk of the statistics community. Since, in Reykjavik, we had discussed the corresponding difficulty of people publishing a longer and “more” statistical paper in a “more” statistical journal, once the central idea was published in a machine learning conference proceeding like NIPS or AISTATS. we had this idea that creating a special fast-track in a mainstream statistics journal for a subset of those papers, using for instance a tailor-made committee in that original conference, or creating an annual survey of the top machine learning conference proceedings rewritten in a more” statistical way (and once again selected by an ad hoc committee) would help, at not too much of a cost for inducing machine learners to make the extra-effort of switching to another style. From there, we enlarged the suggestion to enlist a sufficient number of (diverse) bloggers in each major conference towards producing quick but sufficiently informative entries on their epiphany talks (if any), possibly supported by the conference organisers or the sponsoring societies. (I am always happy to welcome any guest blogger in conferences I attend!)

4 Responses to “bridging the gap between machine learning and statistics”

  1. Hi Christian, I currently edit “Statistical Analysis and Data Mining,” an ASA journal. Its fairly new and is we would be happy to develop a fast track for more statistical papers from AISTATS, UAI, NIPS, etc. In fact we already publish a special issue each year with the SIAM data mining conference. Regards, David

  2. This makes a lot of sense Christian. We’ve been thinking along these lines in the ML community. Unfortunately, as we discussed at AISTATS, many statistics journals are against it. From my perspective, another problem is that many statistics journals are closed access. It would be really great if there was a prestigious stats journal that was open access, like JMLR, which could also help bridge the communitiies keeping information open!

  3. Dan Simpson Says:

    I just spent a (lovely!) week at a Big Data conference in Lofoten and I learnt for the first time what the difference between machine learning and statistics was (I don’t remember, but there was definitely one that was absolutely stated as fact… maybe it had to do with prediction… I’m not sure if stats and ML are distinct, but THEY’RE DEFINITELY DIFFERENT [and stats is better]).

    I’m cynical about these things. I don’t think machine learners need more incentive. If an idea is good, they follow it up, often in a longer-form journal (JML is the classic here, but STCO see some too [or, at least, I’ve reviewed ML style papers for STCO. I didn’t check if it was from a machine learning person or if it was just nomenclature/style])

    But even when they don’t publish in a longer journal (or in stats form), they continue mining the seam. And I don’t think statisticians are naive enough to ignore the ML conference proceedings. But we take them with a grain of salt, because there is a lot of crap there. (The best sign that it’s a good idea seems to be the existence of a follow up paper..)

    Maybe an annotated index (in blog form or otherwise) would be useful?

    I guess a question I have is “are ML people discouraged from statistical publishing?” I mean, they can. ‘They have the technology’

    Is it like the way that some maths departments don’t count publications in science journals? Or is it that the balance of conference vs papers used by ML departments discourages long-form publishing? Or do the stats journals discourage it [because… um … that’s…. bad!]?

    On a related note, the conference seemed to be split between those who thought they had missed the “data science” boat, those who thought it was still being built, and those who had defined themselves to be on it. It was interesting.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.