in which I divulge great nerd secrets

Apr 05, 2006 22:56

it's about math this time
so if you don't want to know )

Leave a comment

Comments 3

(The comment has been removed)

phygelus April 6 2006, 15:42:35 UTC
yeah, I took complex analysis and it's surprising who gets freaked out by that---I don't think most engineers, even EEs, see complex analysis as undergrads at all anymore, which is just wrong---but it won't help you recognize 3sin(z) - 4sin^3(z) as sin(3z). Doing that with complex exponentials is a lot more algebra.

integrating x^x, that's cute. I know I've seen it before, ages ago, but man is that pathological. Silly interview question though.

Reply

(The comment has been removed)

I freely admit I had to google phygelus April 12 2006, 07:02:30 UTC
it's a classic trick question, like integrating sin(x)/x, you have to do it numerically or at least with a series expansion, and you have to pay attention to where it misbehaves. Cute what it does between negative integers, huh?

The wrong way to do it, which is what I would have done in an interview, unprepared, is to rewrite it as e^(x*log(x)) and then try to integrate by parts or something.

There's a cute problem in the beginning of "Numerical Methods That Usually Work" that is pretty much the simplest geometric/trigonometric problem you have to solve numerically. Of course I went roundy-round on it thinking "well I could write the series expansion but there's got to be a closed-form solution" hahahahah

So I guess the moral of the story is, know your series expansions for at least the transcendental functions cold too. Or at least be real ready to go there.

Reply


Leave a comment

Up