Re: [MLUG] N-Queens Problem Solver

[Stephen Montgomery-Smith <EMAIL:PROTECTED>]

Jonathan King wrote:
> On Sat, Apr 19, 2008 at 6:53 AM, Stephen Montgomery-Smith
> <EMAIL:PROTECTED> wrote:
> > > Before i make a real effort I want to be sure because it kinda looks like
> > > maybe you copied my "N" and then typed "n" when you meant "N".
> > >
> > > If by "student," you meant "grad student," then maybe I shouldn't even
> > try, but if you meant high school student, then I probably have some hope of
> > succeeding.
> >
> >  I was an undergraduate.  But it did take me about six months to figure it
> > out.
>
> Wow. And he posts the thing on a Friday evening without telling us
> this, probably just to trying to get us to bite. That was not very
> nice. Let's do the math: took Stephen 6 months to do as a strapping 19
> year old maths major in his calculational prime. Mike and I are
> doddering psychologists in our 40s who have kids and stuff. I estimate
> this could take me, personally, til the twelfth of never.
>
> jking
>
> ps--is there any clever way to use facts about things like
> hypergeometric series here or am I veering off the true path?

I don't know if hypergeometric series would help - I don't know much 
about them - but my guess is that it might be related.

My initial solution was rather complicated, but about six or seven years 
later I spotted a much simpler solution.  Apply the operation f'(y-1) to 
both sides, and use induction.

It is related to the fact that the power series for the inverse function 
to x e^(-x) is something like sum_{n=1}^infty n^(n-1) x^n/n! - this is 
the so called W-Lambert function.



Stephen

_______________________________________________
members mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/members