Author here. I know that treating the L^2 ball as somehow equivalent to binary vectors is invalid - I even say so in the post. If it helps, I explicitly point out the place where the real work for the binary vector would have to go.
The basic idea is I think is to use the central limit theorem on dot(u,d) and dot(p,d) - but unfortunately the CLT does not show that these two dot products are independent.
I am pretty sure the GP was being sarcastic with that list of alternatives (if it just mentioned one of those things, the GP might have just made a mistake, but by naming three controversial things and nothing else, it was likely said sarcastically.
There is nothing sexist or racist about this article, unless you believe that stating a black person is black is somehow offensive towards black people.
Like many people, I hold certain aesthetic preferences. You can call me racist while continuing to prefer blondes to brunettes if it makes you feel good about yourself.
I'm somewhat amused that both women get +1 for great legs and both get 0 for "not letting me work", heh.
It's also interesting that white Svetlana gets +1 for being smart, while black Elise gets +1 for being black but isn't smart. Interesting little Freudian slip (is that the right term?) there, I guess.
I also agree with other commenters that introducing sex/race into this discussion seems completely unnecessary. A better example might be cars, comparing a minivan vs a convertible or something less... politically charged.
Law of small numbers - with only 2 examples, you have to cram lots of possibilities in there. The author envisions two people he might date, and a variety of possible attributes. So he would date someone if they were smart or not, exciting or not, black or not. But probably not someone fat. The whole exercise is an exploration of what the author thinks of himself and has nothing to do with actual women.
The Dirichlet distribution I chose is merely the uniform distribution over the unit simplex. I.e., it means that all possible h-vectors get equal weight.
It's an attempt to choose as uninformative a prior on h as possible.
28 comments
[ 2.1 ms ] story [ 63.7 ms ] threadThe basic idea is I think is to use the central limit theorem on dot(u,d) and dot(p,d) - but unfortunately the CLT does not show that these two dot products are independent.
Couldn't you pick something less controversial, like say, comparing religions, political parties or sports teams?
A "less controversial" topic of comparison would be boxers vs. briefs.
EDIT: I'm bad at sarcasm detection.
It's also interesting that white Svetlana gets +1 for being smart, while black Elise gets +1 for being black but isn't smart. Interesting little Freudian slip (is that the right term?) there, I guess.
I also agree with other commenters that introducing sex/race into this discussion seems completely unnecessary. A better example might be cars, comparing a minivan vs a convertible or something less... politically charged.
It's an attempt to choose as uninformative a prior on h as possible.