• hdu 3221 Brute-force Algorithm(高速幂取模,矩阵高速幂求fib)


    http://acm.hdu.edu.cn/showproblem.php?pid=3221


    一晚上搞出来这么一道题。。Mark。


    给出这么一个程序。问funny函数调用了多少次。

    我们定义数组为所求:f[1] = a,f[2] = b, f[3] = f[2]*f[3]......f[n] = f[n-1]*f[n-2]。相应的值表示也可为a^1*b^0%p。a^0*b^1%p,a^1*b^1%p,.....a^fib[n-3]*b^fib[n-2]%p。即a,b的指数从n=3以后与fib数列一样。


    由于n非常大。fib[n]也想当大。

    a^fib[n]%p能够利用a^fib[n]%p = a^(fib[n]%phi[p]+phi[p])%p进行降幂,条件时fib[n]>=phi[p]。求fib[n]%phi[p]能够构造矩阵。利用矩阵高速幂求fib[n]%phi[p]。


    #include <stdio.h>
    #include <iostream>
    #include <map>
    #include <set>
    #include <list>
    #include <stack>
    #include <vector>
    #include <math.h>
    #include <string.h>
    #include <queue>
    #include <string>
    #include <stdlib.h>
    #include <algorithm>
    #define LL long long
    #define _LL __int64
    #define eps 1e-12
    #define PI acos(-1.0)
    #define C 240
    #define S 20
    using namespace std;
    const int maxn = 110;
    
    struct matrix
    {
        LL mat[3][3];
        void init()
        {
            memset(mat,0,sizeof(mat));
            for(int i = 0; i < 2; i++)
                mat[i][i] = 1;
        }
    } m;
    
    LL a,b,p,n,phi_p;
    LL fib[10000000];
    
    //phi[p]
    LL Eular(LL num)
    {
        LL res = num;
        for(int i = 2; i*i <= num; i++)
        {
            if(num%i == 0)
            {
                res -= res/i;
                while(num%i == 0)
                    num /= i;
            }
        }
        if(num > 1)
            res -= res/num;
        return res;
    }
    //矩阵相乘
    matrix mul_matrix(matrix x, matrix y)
    {
        matrix ans;
        memset(ans.mat,0,sizeof(ans.mat));
        for(int i = 0; i < 2; i++)
        {
            for(int k = 0; k < 2; k++)
            {
                if(x.mat[i][k] == 0) continue;
                for(int j = 0; j < 2; j++)
                {
                    ans.mat[i][j] = (ans.mat[i][j] + x.mat[i][k]*y.mat[k][j])%phi_p;
                }
            }
        }
        return ans;
    }
    //a^t%phi_p
    LL pow_matrix(LL t)
    {
        matrix a,b;
        a.mat[0][0] = a.mat[0][1] = a.mat[1][0] = 1;
        a.mat[1][1] = 0;
        b.init();
        while(t)
        {
            if(t&1)
                b = mul_matrix(a,b);
            a = mul_matrix(a,a);
            t >>= 1;
        }
        return b.mat[0][0];
    }
    //a^t%p
    LL pow(LL a, LL t)
    {
        LL res = 1;
        a %= p;
        while(t)
        {
            if(t&1)
                res = res*a%p;
            a = a*a%p;
            t >>= 1;
        }
        return res;
    }
    //a^fib[t]%p转化为a^(fib[t]%phi[p]+phi[p])%p,fib[t] >= phi[p]。
    LL solve(LL a, LL t)
    {
        fib[0] = 1;
        fib[1] = 1;
        int i;
        for(i = 2; i <= t; i++)
        {
            fib[i] = fib[i-1] + fib[i-2];
            if(fib[i] >= phi_p)
                break;
        }
        if(i <= t) //当满足条件fib[t] >= phi[p]时,进行降幂
        {
            LL c = pow_matrix(t) + phi_p;
            return pow(a,c);
        }
        else
            return pow(a,fib[t]);
    }
    
    int main()
    {
        int test;
        scanf("%d",&test);
        for(int item = 1; item <= test; item++)
        {
            scanf("%lld %lld %lld %lld",&a,&b,&p,&n);
            printf("Case #%d: ",item);
            if(n == 1)
            {
                printf("%lld
    ",a%p);
                continue;
            }
            if(n == 2)
            {
                printf("%lld
    ",b%p);
                continue;
            }
            if(n == 3)
            {
                printf("%lld
    ",a*b%p);
                continue;
            }
            if(p == 1)
            {
                printf("0
    ");
                continue;
            }
            phi_p = Eular(p);
            LL res = solve(a,n-3)*solve(b,n-2)%p;
            printf("%lld
    ",res);
        }
        return 0;
    }
    


  • 相关阅读:
    基于windows的mysql读写分离和amoeba配置
    windows环境下mysql主从配置
    Mysql主从报错锦集
    Nginx流控
    windows_windows下的rsync
    Kafka+ZooKeeper高可用集群部署
    通过Nginx、Consul、Upsync实现动态负载均衡和服务平滑发布
    Redis性能指标监控
    MongoDB用户权限管理
    ZooKeeper原理详解及常用操作
  • 原文地址:https://www.cnblogs.com/yxwkf/p/5113611.html
Copyright © 2020-2023  润新知