deep learning in Toulouse [post-doc position]

An opening for an ERC post-doc position on Bayesian deep learning with Cédric Févotte in Toulouse.

the “myth of the miracle machine”

In what appears to be a regular contribution of his to Nature, Daniel Sarewitz recently wrote a “personal take on events” that I find quite reactionary, the more because it comes from an academic. And I wonder why Nature chose to publish his opinion piece. Every other month! The arguments of the author is that basic science should be defunded in favour of “use-inspired” research, “mission oriented” programmes, “societal needs and socially valuable knowledge”… The reason being that it is a better use of public money and that scientists are just another interest group that should not be left to its own device. This is not a new tune, calls to cut down funding fundamental research emerge regularly as an easily found culprit for saving “taxpayer money”, and it is the simplest mean of rejecting a research proposal by blaming its lack of clear applicability. Of course, when looking a bit wider, one can check this piece bemoaning the Democrat inclinations of most scientists. Or that one that science should sometimes give way to religion. With the definitive argument that, for most people, the maths behind scientific models are so complex that they must turn to an act of faith… Yes, I do wonder at Nature providing Sarewitz with such a wide-ranging tribune.

OBayes 17 travel support

The OBayes 17 conference in Austin, Texas, next December is getting nearer! This post is to advertise for the availability of a dozen travel grants for junior investigators, as detailed on the webpage of the conference. One of those grants will even become an ISBA New Researchers Travel Award for the event! This comes on top of registration and accommodation being quite reasonable, thanks to Peter Mueller’s efforts, and hence makes this conference most affordable and attractive for young researchers. Apply now!!!

why I cannot stand committees’ email debates…

I generally dislike committee work for the amount of useless [or low signal-to-noise] email it induces. In a conference committee I am was part of, a debate started on whether or not we should accept sponsorship from a corporate sponsor and nominate a speaker for the associated lecture (named after a prestigious scientist, not after the sponsor). Some members of the committee went viral about this and inundated us with requests for action and for taking position, to such a level that I eventually reacted with the following email to the initiator of the virus:

Dear Zeno of Elea,

While I stayed away from a debate I was not in the least interested in, the latest emails were so aggressive and borderline insulting that I felt compelled to rise to the bait and to add to an already high level of static noise. When I was asked to join this committee, it was to nominate plenary speakers for a conference, not to debate about the sorry state of the World, nor to hear about your quibble on whether or not accepting sponsor money from one corporate source or another is evil.

On a general basis, I am not interested in being forced into debates in this low-content high-noise format and I cannot stand the invasion of my mailbox by a flow of emails that has turned viral. I hope this point contributes to enlighten you as to why some members of the committee did not reply to earlier emails or to the call for a doodle vote. I am simply not interested. And I do not think any of this gesticulation will have any impact on the sorry state of the World. So please leave me out of this debate.

My incensed reaction is not about the sponsorship issue (which I think is terribly trivial) but (a) about the holier-than-thou attitude that considers that everyone should take a stand, chose one side (with an obvious choice), and be ashamed of one’s lack of involvment, and (b) about using collective emails to enforce this strategy. I am a strong believer in Arrow’s theorem and email debates are adding another magnitude of impossibility to committee decisions.

Fellowships in Stat only: good news?!

As reported in Nature newsblog, the UK funding body, EPSRC (Engineering and Physical Sciences Research Council), “has scrapped fellowships in all but two areas of mathematical sciences, namely statistics and applied probability”. This decision may sound like a bonanza for statisticians and applied probabilists, however, when thinking about it a bit more widely, it is close to a disaster. Choosing to fund fellowships only in a narrow subset of the field is indeed unfair, unwise, and inefficient. Unfair because the topics were chosen w/o consultation with mathematicians. It could have been numerical analysis or cryptography instead. In which case it would have impacted statisticians and applied probabilists as well. Thus, top UK statisticians like Peter Donnelly and Peter Green rightly signed a protest letter along colleagues from other mathematical fields. (Maybe the RSS has likewise reacted. I have not seen it.) Unwise, because, as noted in the letter sent a week ago by twenty-five top UK mathematicians to their Prime Minister, cutting funds in most of mathematics will mean that most UK PhD students will leave the UK to get fellowships abroad. With a fair chance of never returning. (Maybe a bonanza for France? Not really, either, as the funding has not increased here and the current French PhDs need to be funded as well. Even though they most often get hired within a few months of their defense. Or leave for a postdoc abroad…) Inefficient, because the decision is taken without prior notice and cannot expect to impact the area of research of future PhD’s. Nor does it bring a solution for the future of current PhD’s in Not!{statistics and applied probability}…