hdu_2925_Musical Chairs_201311121643

Musical Chairs

Time Limit: 2000/1000 MS (Java/Others)    Memory Limit: 32768/32768 K (Java/Others) Total Submission(s): 898    Accepted Submission(s): 509

Problem Description
In the traditional game of Musical Chairs, N + 1 children run around N chairs (placed in a circle) as long as music is playing. The moment the music stops, children run and try to sit on an available chair. The child still standing leaves the game, a chair is removed, and the game continues with N children. The last child to sit is the winner.
In an attempt to create a similar game on these days' game consoles, you modify the game in the following manner: N Children are seated on N chairs arranged around a circle. The chairs are numbered from 1 to N . Your program pre-selects a positive number D . The program starts going in circles counting the children starting with the first chair. Once the count reaches D , that child leaves the game, removing his/her chair. The program starts counting again, beginning with the next chair in the circle. The last child remaining in the circle is the winner.
For example, consider the game illustrated in the figure above for N = 5 and D = 3 . In the figure, the dot indicates where counting starts and × indicates the child leaving. Starting off, child #3 leaves the game, and counting restarts with child #4. Child #1 is the second child to leave and counting restart with child #2 resulting in child #5 leaving. Child #2 is the last to leave, and child #4 is the winner. Write a program to determine the winning child given both N and D .
 
Input
Your program will be tested on one or more test cases. Each test case specifies two positive integers N and D on a single line, separated by one or more spaces, where N, D < 1, 000, 000 .
The last line of the input file contains two 0's and is not part of the test cases.
 
Output
For each test case, write the winner using the following format:
N D W
Where N and D are as above, is a space character, and W is the winner of that game.
 
Sample Input
5 3
7 4
0 0
 
Sample Output
5 3 4
7 4 2
 
 
 1 #include <stdio.h>
 2 int main()
 3 {
 4     int n,d;
 5     while(scanf("%d %d",&n,&d),n||d)
 6     {
 7         int i,w=0;
 8         for(i=2;i<=n;i++)
 9         w=(w+d)%i;
10         printf("%d %d %d\n",n,d,w+1);
11     }
12     return 0;
13 }

 

约瑟夫环问题:

http://baike.baidu.com/link?url=DLfL-nqyyQmoxRRs7ZnN6NxV2wAX1FDDT2I0kboKxbSK95NzWRx2ITNsgjrUw13W

数学方法

无论是用链表实现还是用数组实现都有一个共同点:要模拟整个游戏过程,不仅程序写起来比较烦,而且时间复杂度高达O(nm),当n,m非常大(例如上百万,上千万)的时候,几乎是没有办法在短时间内出结果的。我们注意到原问题仅仅是要求出最后的胜利者的序号,而不是要读者模拟整个过程。因此如果要追求效率,就要打破常规,实施一点数学策略。
为了讨论方便,先把问题稍微改变一下,并不影响原意:
问题描述:n个人(编号0~(n-1)),从0开始报数,报到m-1的退出,剩下的人继续从0开始报数。求胜利者的编号。
我们知道第一个人(编号一定是(m-1)%n) 出列之后,剩下的n-1个人组成了一个新的约瑟夫环(以编号为k=m%n的人开始):
k k+1 k+2 ... n-2,n-1,0,1,2,... k-2
并且从k开始报0。
我们把他们的编号做一下转换:
k --> 0
k+1 --> 1
k+2 --> 2
...
...
k-3 --> n-3
k-2 --> n-2
序列1:0,1,2,3 … n-2,n-1
序列2:0,1,2,3 … k-2,k,…,n-2,n-1
序列3:k,k+1,k+2,k+3,…,n-2,n-1,0,1,2,3,…,k-2,
序列4:0,1,2,3 …,5,6,7,8,…,n-3,n-2
变换后就完完全全成为了(n-1)个人报数的子问题,假如我们知道这个子问题的解:例如x是最终的胜利者,那么根据上面这个表把这个x变回去不刚好就是n个人情况的解吗?!!变回去的公式很简单,相信大家都可以推出来:
∵ k=m%n;
∴ x' = x+k = x+ m%n ; 而 x+ m%n 可能大于n
∴x'= (x+ m%n)%n = (x+m)%n
得到 x‘=(x+m)%n
如何知道(n-1)个人报数的问题的解?对,只要知道(n-2)个人的解就行了。(n-2)个人的解呢?当然是先求(n-3)的情况 ---- 这显然就是一个倒推问题!好了,思路出来了,下面写递推公式:
令f表示i个人玩游戏报m退出最后胜利者的编号,最后的结果自然是f[n].
递推公式:
f[1]=0;
f[i]=(f[i-1]+m)%i; (i>1)
有了这个公式,我们要做的就是从1-n顺序算出f的数值,最后结果是f[n]。因为实际生活中编号总是从1开始,我们输出f[n]+1。
我们输出f[n]由于是逐级递推,不需要保存每个,程序也是异常简单:(注意编号是0 -- n-1)
#include <stdio.h>
int main(void)
int n,m,i,s=0;
printf ("N M = ");
scanf("%d%d",&n,&m);
for (i=2; i<=n; i++)
s=(s+m)%i;
printf ("The winner is %d\n",s);
return 0 ;
时间复杂度为O(n),相对于模拟算法已经有了很大的提高。算n,m等于一百万,一千万的情况不是问题了。可见,适当地运用数学策略,不仅可以让编程变得简单,而且往往会成倍地提高算法执行效率。
参照上面提供的思路,我认为可以类似的得到一个更易于明白的方法,设有(1,2,3,……,k-1,k,k+1,……,n)n个数,当k出列时,那么有
k+1 -->1
k+2 -->2
...
...
n -->n-k
1 -->n-k+1
...
...
k-1 -->n-1
由上面一组式子可以推出,若知道新产生的n-1个数中某个数x,那么很显然可以推出x在原数列里的位置,即x‘=(x+k)%n,由此,我们可以得到一个递推公式
f[1]=1
f[n]=(f[n-1]+k)%n (n>1)
如果你认为上式可以推出约瑟夫环问题的解,很不幸,你错了,上面的递推公式中,在某种情况下,f[n-1]+k会整除n,如n=2,k=3,这时我们需要对上式进行修正,
f[n]=(f[n-1]+k)%n;if(f[n]==0)f[n]=n;
问题得解。程序代码如下:
#include<stdio.h>
int main()
int n,k,s=1;
scanf("%d%d",&n,&k);
for(int i=2;i<=n;i+=1)
s=(s+k)%i;
if(s==0)s=i;
printf("ans=%d\n",s);
return 0;
 
 
 
posted @ 2013-11-12 16:52  龙腾四海365  阅读(252)  评论(0编辑  收藏  举报