Friday, October 26, 2007

Oh. My. God.

Eliezer Yudkowsky is super duper smart. He's 100 times smarter than me, which is probably why he intended this for primary school kids. In my experience, clever people often know they're clever but massively overestimate what other people should know, which is actually very stupid. It is so pathetically out of touch with reality I don't know how he can excuse it. Most types of intelligence correlate quite well with each other (although this doesn't fit the image of the socially inept mathematician. I think that's a special case), but this is a genuine and terrible stupidity so don't laugh it off you smug bastard. Will my dear, unsmug readers challenge the claim that smart people actually enjoy demonstrating their superiority in this way? Ah, the age old question, is person X evil or stupid? Here's an easily pasted section.

The right side of Bayes' Theorem is derived from the left side through these steps:

p(AX) =

p(AX)

p(AX) =

p(X&A)
p(X)

p(AX) =

p(X&A)
p(X&A) + p(X&~A)

p(AX) =

p(XA)*p(A)
p(XA)*p(A) + p(X~A)*p(~A)


The first step, p(AX) to p(X&A)/p(X), may look like a tautology. The actual math performed is different, though. p(AX) is a single number, the normalized probability or frequency of A within the subgroup X. p(X&A)/p(X) are usually the percentage frequencies of X&A and X within the entire sample, but the calculation also works if X&A and X are absolute numbers of people, events, or things. p(cancerpositive) is a single percentage/frequency/probability, always between 0 and 1. (positive&cancer)/(positive) can be measured either in probabilities, such as 0.008/0.103, or it might be expressed in groups of women, for example 194/2494. As long as both the numerator and denominator are measured in the same units, it should make no difference.

PRIMARY SCHOOL!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

5 comments:

cristi said...

are you joking? is he joking? I'm not sure, but i don't get it. granted i only read it once and very quickly, and my mind wandered after the second line of math, but is that really easy?????

cristi said...

never mind, i misread second sentence (i read "underestimate" instead of "overestimate"...that changes everything. glad i'm not supposed to get it.

Tracy Leigh said...

He is obviously out of touch with reality, since he thinks that knowing Bayesian theory is cool and that all your primary school friends will be wearing t-shirts with Bayesian theory on it!

Nonetheless, I think it's actually quite a useful article (not that I've finished reading it).

mutt said...

it is good and he does realise now that its good for grown ups like us. I also think he was kinda kidding when he talks about how cool it is.

I still think he's an arrogant prick though.

mutt said...

looks like i'm going for the evil option. I still think stupid is almost as likely.