Hacker News Reader

Comment by eli_gottlieb | Hacker News Reader

Statistics is formalized philosophy of science, so of course scientists outside the stats departments wanted "easy tests". Arguably, a lot of the new applications of computational Bayesian methods in the natural science are the same statistical witch-doctoring with a bit more rigor.

(Ok, I do computational Bayes for a living with natural scientists, so I'm inherently biased to view our work as shakier than it really is.)

View on HN

'Statistics is formalized philosophy of science' -- care to expand? Or provide supplementary literature?

He’s likely referring to the fact that the rules of probabilistic reasoning can be shown to be the only rules of reasoning that yield correct inferences from data. Bayesian statistics is in that sense the “logic of science”. The first chapter of E.T. Jaynes’s book with that subtitle [0] is a good introduction to these ideas if you haven’t seen them before.

0. E.T. Jaynes (2003), “Probability Theory: the Logic of Science,” http://bayes.wustl.edu/etj/prob/book.pdf [PDF]

As a matter of fact, Jaynes is a good reference even though I didn't have him in mind at the time. I was more referring to the general impression I get reading a lot of statistics textbooks and papers, which is that they're trying to numerically formalize methodologies appropriate to experimental sciences. You end up with a fair amount of papers co-written between phil-of-science scholars and statisticians.

Aside: What do you do for a living?

LinkedIn in his HN user page.

> Variational Bayesian methods and deep learning for affective neuroscience, using probabilistic programming systems ProbTorch and Pyro.

It does sound like he's in for a rough time, epistemologically speaking.

You'd be surprised how easily the witch-doctory passes for real science ;-)! Kidding, kidding, our empirical side actually has to be at least as rigorous as the other statistical methods used in neuroscience, usually more so. That's what they keep us around for!

Twenty years ago I would have been surprised. :-)

Reply on news.ycombinator.com