Subject: | |
From: | |
Reply To: | Classification, clustering, and phylogeny estimation |
Date: | Mon, 10 May 2004 21:55:01 +0100 |
Content-Type: | TEXT/PLAIN |
Parts/Attachments: |
|
|
Hi All ,
does anybody know an upper bound for the probability of the
largest Voronoi cell?
More precisely, let x_1 ... x_n be drawn randomly and independently
according
to a distribution P on R^d. I'm looking for an upper bound of
E ( max_{i=1..n} P(B_i) )
where B_i is the Voronoi cell containing x_i, i.e.
B_i={x: x_i is the nearest neighbour of x among x_1,...,x_n}
I guess this should be O(1/n) ... but not sure.
thanks,
Daniil
|
|
|