UVA - 1456(cellular network 概率dp)

首先，要把概率从小到大排序，原因很简单，在期望表达式中同样的决策方案人数越多的项占有的比重越小越好；

然后直接dp即可；d[ i ][ j ] 代表到i位置已经分了几组剩下分配的最大期望；

#include <cstdio>
#include <map>
#include <cmath>
#include <iostream>
#include <cstring>
#include <algorithm>
#include <vector>
using namespace std;
const int inf  = 1000;
const int maxn = 105;
double d[maxn][maxn],w[maxn],sum[maxn];
bool vis[maxn][maxn];
int n,m;
double Sum(int x,int y){ return w[y]-w[x-1];}
double dp(int i,int j){
  if(vis[i][j]) return d[i][j];
  vis[i][j] = true;
  if(i==n&&j<m) return d[i][j]=inf;
  if(j==m-1){
     return d[i][j] = n*Sum(i+1,n);
  }
  d[i][j]=inf;
  for(int k=i+1;k<=n;k++){
     d[i][j]=min(d[i][j],dp(k,j+1)+k*Sum(i+1,k));
  }
  return d[i][j];
}
int cmp(const double a,const double b){
  return a > b;
}
int main()
{
    int T;
    scanf("%d",&T);
    while(T--){
       scanf("%d %d",&n,&m);
       int he=0;
       for(int i=1;i<=n;i++){
          scanf("%lf",&sum[i]);
          he+=sum[i];
       }
       w[0]=0;
       for(int i=1;i<=n;i++){
         w[i]=sum[i]/he;
       }
       sort(w+1,w+n+1,cmp);
       for(int i=1;i<=n;i++){
         w[i]=w[i]+w[i-1];
       }
       memset(vis,false,sizeof(vis));
       printf("%.4lf\n",dp(0,0));
    }
    return 0;
}