Fibonacci(矩阵快速幂 - 递推式构造 + 变种)

POJ 3070 Fibonacci

题目链接：http://poj.org/problem?id=3070

题意：求F_n % 1000是多少？(公式题目已给出)
此题的公式就是F_n = F_n-1 + F_n-2，我们令:
  $X_{n-1}=\left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\end{array}\right]$  $X_n=\left[\begin{array}{c}{F}_n\\ F_{n-1}\end{array}\right]$
那么我们需要找到的矩阵 A 就满足：
      $A\cdot X_{n-1}=X_n$
又因为：
    $\{\begin{array}{l}{F}_n=F_{n-1}+F_{n-2}\\ F_{n-1}=F_{n-1}+0\cdot F_{n-2}\end{array}$
所以：
    $A=\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]$
即：
    $\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]$ $\cdot$ $\left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\end{array}\right]$$=\left[\begin{array}{c}{F}_n\\ F_{n-1}\end{array}\right]$
综上所述：
    $\left[\begin{array}{c}{F}_n\\ F_{n-1}\end{array}\right]=\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]\cdot \left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\end{array}\right]={\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]}^2\cdot \left[\begin{array}{c}{F}_{n-2}\\ F_{n-3}\end{array}\right]\cdot \cdot \cdot \cdot ={\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]}^{n-1}\cdot \left[\begin{array}{c}{F}_1\\ F_0\end{array}\right]$


我们用矩阵快速幂求出${\left[\begin{array}{c}1,1\\ 0,0\end{array}\right]}^{n-1}$后，结果矩阵的第一行与上面公式中最后的$\left[\begin{array}{c}{F}_1\\ F_0\end{array}\right]$做矩阵乘法即是最终结果。

#include<cstdio>
#include<cstring>
#include<iostream>
#include<algorithm>
#define mod 10000
using namespace std;

typedef long long LL;
const int Max_n=100005;

struct Mat{
    int mat[2][2];
};

Mat mul(Mat a,Mat b){
    Mat ans;
    for(int i=0;i<2;i++){
        for(int j=0;j<2;j++){
            ans.mat[i][j]=0;
            for(int k=0;k<2;k++)
                ans.mat[i][j]+=(a.mat[i][k]*b.mat[k][j])%mod;
                ans.mat[i][j]%=mod;
        }
    }
    return ans;
}

Mat q_pow(Mat a,int b){
    Mat ans={1,0,0,1};
    Mat res=a;
    while(b){
        if(b&1) ans=mul(ans,res);
        res=mul(res,res);
        b>>=1;
    }
    return ans;
}
int main(){
    int n;
    while(~scanf("%d",&n)&&n!=-1){
        Mat a={1,1,1,0};
        Mat ans={0,0,0,0};
        ans=q_pow(a,n); 
        printf("%d\n",ans.mat[0][1]);
    }
    return 0;
}

2018年湘潭大学程序设计竞赛 G 又见斐波那契

题目地址:https://ac.nowcoder.com/acm/contest/105/G?&headNav=www&headNav=acm
本题需要注意的点就是矩阵的数据类型,用int中间运算会爆掉,所以构造矩阵的时候需要用 ll
思路：与上面题目套路相同
  $X_{n-1}=\left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\\ i^3\\ i^2\\ i\\ 1\end{array}\right]$  $X_n=\left[\begin{array}{c}{F}_n\\ F_{n-1}\\ (i+1{)}^3\\ (i+1{)}^2\\ (i+1)\\ 1\end{array}\right]$  $A\cdot X_{n-1}=X_n$


  $\left[\begin{array}{c}X,X,X,X,X,X\\ X,X,X,X,X,X\\ X,X,X,X,X,X\\ X,X,X,X,X,X\\ X,X,X,X,X,X\\ X,X,X,X,X,X\end{array}\right]$ $\cdot$ $\left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\\ (i+1{)}^3\\ (i+1{)}^2\\ i+1\\ 1\end{array}\right]$$=\left[\begin{array}{c}{F}_n\\ F_{n-1}\\ i^3\\ i^2\\ i\\ 1\end{array}\right]$


$\{\begin{array}{l}{F}_n=1\cdot F_{n-1}+1\cdot F_{n-2}+1\cdot i^3+1\cdot i^2+1\cdot i+1\cdot 1\\ F_{n-1}=1\cdot F_{n-1}+0\cdot F_{n-2}+0\cdot i^3+0\cdot i^2+0\cdot i+0\cdot 1\\ (i+1{)}^3=0\cdot F_{n-1}+0\cdot F_{n-2}+1\cdot i^3+3\cdot i^2+3\cdot i+1\cdot 1\\ (i+1{)}^2=0\cdot F_{n-1}+0\cdot F_{n-2}+0\cdot i^3+1\cdot i^2+2\cdot i+1\cdot 1\\ i+1=0\cdot F_{n-1}+0\cdot F_{n-2}+0\cdot i^3+0\cdot i^2+1\cdot i+1\cdot 1\\ 1=0\cdot F_{n-1}+0\cdot F_{n-2}+0\cdot i^3+0\cdot i^2+0\cdot i+1\cdot 1\end{array}$ $=>$   $A=\left[\begin{array}{c}1,1,1,1,1,1\\ 1,0,0,0,0,0\\ 0,0,1,3,3,1\\ 0,0,0,1,2,1\\ 0,0,0,0,1,1\\ 0,0,0,0,0,1\end{array}\right]$


$\left[\begin{array}{c}{F}_n\\ F_{n-1}\\ (i+1{)}^3\\ (i+1{)}^2\\ i+1\\ 1\end{array}\right]=\left[\begin{array}{c}1,1,1,1,1,1\\ 1,0,0,0,0,0\\ 0,0,1,3,3,1\\ 0,0,0,1,2,1\\ 0,0,0,0,1,1\\ 0,0,0,0,0,1\end{array}\right]\cdot \left[\begin{array}{c}{F}_{n-1}\\ F_{n-2}\\ i^3\\ i^2\\ i\\ 1\end{array}\right]={\left[\begin{array}{c}1,1,1,1,1,1\\ 1,0,0,0,0,0\\ 0,0,1,3,3,1\\ 0,0,0,1,2,1\\ 0,0,0,0,1,1\\ 0,0,0,0,0,1\end{array}\right]}^2\cdot \left[\begin{array}{c}{F}_{n-2}\\ F_{n-3}\\ (i-1{)}^3\\ (i-1{)}^2\\ i-1\\ 1\end{array}\right]=\cdot \cdot \cdot \cdot ={\left[\begin{array}{c}1,1,1,1,1,1\\ 1,0,0,0,0,0\\ 0,0,1,3,3,1\\ 0,0,0,1,2,1\\ 0,0,0,0,1,1\\ 0,0,0,0,0,1\end{array}\right]}^{n-1}\cdot \left[\begin{array}{c}{F}_1\\ F_0\\ 2^3\\ 2^2\\ 2\\ 1\end{array}\right]$


最后求出$A^{n-1}$后，与公式中最后的行列式相乘即可。

#include<cstdio>
#include<cstring>
#include<iostream>
#include<algorithm>
using namespace std;

typedef long long LL;
const int mod=1000000007;
const int Max_n=500005;
int x[6]={1,0,8,4,2,1};

struct Mat{
    LL mat[6][6];
};

Mat mul(Mat a,Mat b){
    Mat z;
    for(int i=0;i<6;i++){
        for(int j=0;j<6;j++){
            z.mat[i][j]=0;
            for(int k=0;k<6;k++){
                z.mat[i][j]+=(a.mat[i][k]*b.mat[k][j])%mod;
                z.mat[i][j]%=mod;
            }
        }
    }
    return z;
}
Mat q_pow(Mat a,LL b){
    Mat ans={
        1,0,0,0,0,0,
        0,1,0,0,0,0,
        0,0,1,0,0,0,
        0,0,0,1,0,0,
        0,0,0,0,1,0,
        0,0,0,0,0,1,
    };
    Mat res=a;
    while(b){
        if(b&1) ans=mul(ans,res);
        res=mul(res,res);
        b>>=1;
    }
    return ans;
}

int main(){
    Mat a={
        1,1,1,1,1,1,
        1,0,0,0,0,0,
        0,0,1,3,3,1,
        0,0,0,1,2,1,
        0,0,0,0,1,1,
        0,0,0,0,0,1
    };
    int t;
    scanf("%d",&t);
    while(t--){
        LL n;
        scanf("%lld",&n);
        if(n==1){
            printf("1\n");
            continue;
        }
        if(n==0){
            printf("0\n");
            continue;
        }
        Mat ans=q_pow(a,n-1);
        int res=0;
        for(int i=0;i<6;i++){
            res+=(ans.mat[0][i]*x[i])%mod;
            res%=mod;
        }
        printf("%d\n",res);
    }
    return 0;
}