珍珠湾ART

标题: 空间中的球面 [打印本页]

作者: fzy    时间: 2005-10-11 21:31
标题: 空间中的球面

难度:++到+++
空间中N个球面最多能把空间分成多少部分?
www.ddhw.com

 

作者: 野 菜 花    时间: 2005-10-11 22:15
标题: 回复:空间中的球面

f(N)=1/3*N*(N^2-3N+8)
www.ddhw.com

 

作者: fzy    时间: 2005-10-11 23:00
标题: No detail,no credit

  No detail,no credit





作者: 野 菜 花    时间: 2005-10-11 23:10
标题: I learn from you guys, you sometimes only give

answer too. www.ddhw.com
 
Let F(N)=a*N^3+b*N^2+c*N+d
Use F(1)=2, F(2)=4, F(3)=8, F(4)=16
and undetermine coefficient method, solve the system for a,b,c,d
www.ddhw.com

 

作者: fzy    时间: 2005-10-11 23:20
标题: You cannot assume it is a polynomial [:((][:X]

  You cannot assume it is a polynomial





作者: 野 菜 花    时间: 2005-10-11 23:44
标题: I know it from a theorem. I like to see your nicer

solution.
www.ddhw.com

 

作者: 乱弹    时间: 2005-10-12 03:11
标题: Please tell me the thm.. Thanks [@};-][@};-][@};-]

  Please tell me the thm.. Thanks





作者: 野 菜 花    时间: 2005-10-12 04:16
标题: 回复:Please tell me the thm.. Thanks [@};-][@};-][@}

I don't remember what the theorem exactly says, I only got this conclusion from it (but don't take it seriously, maybe my impression is not accurate) :
 
For a k-dim space, let F(N) be the max number of parts that N (k-1)-dim spheres divide the k-dim space into, then F(N) is a polynomial of N and F(N)=O(N^k).
 
In fact, it is true when k=1,2,3
www.ddhw.com

 

作者: 乱弹    时间: 2005-10-12 04:35
标题: 查了一下, 没查到权威的, 不过你说的是对的。[>:D<]

http://www.research.att.com/cgi-bin/access.cgi/as/njas/sequences/eisA.cgi?Anum=A046127
 
 
Formula:   a(n)=n(n^2-3n+8)/3 (n>0).
           n hyperspheres divide R^k into at most C(n-1,k) + Sum_{i=0..k} C(n,i)
              regions.
www.ddhw.com

 

作者: husonghu    时间: 2005-11-10 03:16
标题: 此题标准答案已由fzy给出。有兴趣者请见坛上另贴。

  此题标准答案已由fzy给出。有兴趣者请见坛上另贴。









欢迎光临 珍珠湾ART (http://66.160.158.134/) Powered by Discuz! X3