UVa1374 Power Calculus (IDA*)

题目链接

Starting with x and repeatedly multiplying by x, we can compute x³¹ with thirty multiplications: The operation of squaring can appreciably shorten the sequence of multiplications. The following is a way to compute x³¹ with eight multiplications:(略)
This is not the shortest sequence of multiplications to compute x³¹. There are many ways with only seven multiplications. The following is one of them:(略)
There however is no way to compute x³¹ with fewer multiplications. Thus this is one of the most efficient ways to compute x³¹ only by multiplications. If division is also available, we can find a shorter sequence of operations. It is possible to compute x³¹ with six operations (five multiplications and one division):(略)
This is one of the most efficient ways to compute x³¹ if a division is as fast as a multiplication. Your mission is to write a program to find the least number of operations to compute xⁿ by multiplication and division starting with x for the given positive integer n. Products and quotients appearing in the sequence of operations should be x to a positive integer’s power. In other words, x⁻³, for example, should never appear.

Input
The input is a sequence of one or more lines each containing a single integer n. n is positive and less than or equal to 1000. The end of the input is indicated by a zero.

Output
Your program should print the least total number of multiplications and divisions required to compute xⁿ starting with x for the integer n. The numbers should be written each in a separate line without any superfluous characters such as leading or trailing spaces.

1.题意不再赘述，刚开始想到了快速幂，想在快速幂的过程中求出最小的步数，但实际上快速幂并不是择优选择的，而是从x¹每次开方，当n&1时就乘一次x，实际上x的计算次数还是比较多的

2.看分析说是IDA*便开始构思代码。估价函数自己没想出来，但是看LRJ那么写却可以看懂（还是要多独立思考）。我们知道，每深入搜索一层，那么可以得到该层的最大数值是2^当前层数，那么我们设置了最大搜索层数，如果当前的最大数值为x，x*2^maxd-d<n，代表搜完规定层数的最大值仍然小于n，那么就剪枝（实际上我们发现一开始如果2^maxd<n就不会搜索下去直接返回）。但是如何实现每一次搜索得到的序列集合呢，而且我们又要将集合中的元素都相互加减，那么我觉得需要数组将每个集合的存起来，还要用vis判重，但是实际上代码不许要写那么多

3.我们只用一个数组a记录到当前数x，那么我们每次只用新加入的数x对之前数组中所有的元素加减再dfs，那么可以优化很多dfs。例如刚开始只有{1}，那么1对数组中所有元素加减，接下来加入了2，那么2在对{1,2}加减，这时的1就不需要再加减了（负数是需要返回的）

#include <iostream>

using namespace std;
int n,ans;
int a[1005];

bool dfs(int cur,int x,int maxd){
    if((x<<(maxd-cur))<n) return false;
    if(x<=0 || cur>maxd) return false;
    if(x==n || (x<<(maxd-cur))==n){ ans=cur; return true; };
    a[cur]=x;
    for(int i=0;i<=cur;i++){
        if(dfs(cur+1,x+a[i],maxd)) return true;
        if(dfs(cur+1,x-a[i],maxd)) return true;
    }
    return false;
}

int main()
{
    while(scanf("%d",&n) && n){
        for(int i=0;;i++){  //这里需要从0开始，没注意WA了一次
            if(dfs(0,1,i)){
                printf("%d\n",i);
                break;
            }
        }
    }
    return 0;
}