洛谷传送门
Codeforces传送门
题目描述
For an array b b of length we define the function f f as , where ⊕ ⊕ is bitwise exclusive OR.
For example, f(1,2,4,8)=f(1⊕2,2⊕4,4⊕8)=f(3,6,12)=f(3⊕6,6⊕12)=f(5,10)=f(5⊕10)=f(15)=15 f ( 1 , 2 , 4 , 8 ) = f ( 1 ⊕ 2 , 2 ⊕ 4 , 4 ⊕ 8 ) = f ( 3 , 6 , 12 ) = f ( 3 ⊕ 6 , 6 ⊕ 12 ) = f ( 5 , 10 ) = f ( 5 ⊕ 10 ) = f ( 15 ) = 15
You are given an array a a and a few queries. Each query is represented as two integers and r r . The answer is the maximum value of on all continuous subsegments of the array al,al+1,…,ar a l , a l + 1 , … , a r .
输入输出格式
输入格式:
The first line contains a single integer n n ( ) — the length of a a .
The second line contains integers a1,a2,…,an a 1 , a 2 , … , a n ( 0≤ai≤230−1 0 ≤ a i ≤ 2 30 − 1 ) — the elements of the array.
The third line contains a single integer q q ( ) — the number of queries.
Each of the next q q lines contains a query represented as two integers , r r ( ).
输出格式:
Print q q lines — the answers for the queries.
输入输出样例
输入样例#1:
3
8 4 1
2
2 3
1 2
输出样例#1:
5
12
输入样例#2:
6
1 2 4 8 16 32
4
1 6
2 5
3 4
1 2
输出样例#2:
60
30
12
3
说明
In first sample in both queries the maximum value of the function is reached on the subsegment that is equal to the whole segment.
In second sample, optimal segment for first query are , for second query — [2,5] [ 2 , 5 ] , for third — [3,4] [ 3 , 4 ] , for fourth — [1,2] [ 1 , 2 ] .
题目大意
给你一个序列, 一次操作可以将长为 n n 一段序列的相邻元素两两异或, 得到个值(顺序不变), 如此操作 n−1 n − 1 次可以得到一个值。 现给出 q q 组询问, 每次询问在范围内操作其中子序列可以得到最大值是多少。、
解题分析
首先很容易看出
[l,r]
[
l
,
r
]
的序列每个元素异或的次数和
(r−l+1i−l+1)
(
r
−
l
+
1
i
−
l
+
1
)
对应, 这样我们就有了一个
O(n3)
O
(
n
3
)
的优秀做法。
那么怎么优化? O(n3) O ( n 3 ) 的原因是我们没法直接转移, 必须每次暴力算。 那么我们可以发现, 设 dp[i][j] d p [ i ] [ j ] 表示 [i,j] [ i , j ] 的最终的值, 那么 dp[i][j]=dp[i+1][j]⊕dp[i][j−1] d p [ i ] [ j ] = d p [ i + 1 ] [ j ] ⊕ d p [ i ] [ j − 1 ] , 这样我们记忆化一下, 可以得到 O(n2) O ( n 2 ) 的做法。
代码如下:
#include <cstdio>
#include <cstring>
#include <cmath>
#include <algorithm>
#include <cctype>
#include <cmath>
#define R register
#define IN inline
#define W while
#define gc getchar()
#define MX 5050
template <class T>
IN void in(T &x)
{
x = 0; R char c = gc;
for (; !isdigit(c); c = gc);
for (; isdigit(c); c = gc)
x = (x << 1) + (x << 3) + c - 48;
}
int n, q;
int dp[MX][MX], ans[MX][MX], val[MX];
int DP (R int lef, R int rig)
{
if(lef == rig) return val[lef];
if(~dp[lef][rig]) return dp[lef][rig];
return dp[lef][rig] = DP(lef + 1, rig) ^ DP(lef, rig - 1);
}
int getans(R int lef, R int rig)
{
if(lef == rig) return val[lef];
if(~ans[lef][rig]) return ans[lef][rig];
return ans[lef][rig] = std::max(std::max(getans(lef + 1, rig), getans(lef, rig - 1)), DP(lef, rig));
}
int main(void)
{
std::memset(dp, -1, sizeof(dp)); std::memset(ans, -1, sizeof(ans));
int a, b;
in(n);
for (R int i = 1; i <= n; ++i) in(val[i]);
in(q);
W(q--)
{
in(a), in(b);
printf("%d\n", getans(a, b));
}
}