F. Bulbo

Time Limit: 1 Sec

Memory Limit: 256 MB

题目连接

http://codeforces.com/contest/575/problem/F

Description

Bananistan is a beautiful banana republic. Beautiful women in beautiful dresses. Beautiful statues of beautiful warlords. Beautiful stars in beautiful nights.

In Bananistan people play this crazy game – Bulbo. There’s an array of bulbs and player at the position, which represents one of the bulbs. The distance between two neighboring bulbs is 1. Before each turn player can change his position with cost |posnew - posold|. After that, a contiguous set of bulbs lights-up and player pays the cost that’s equal to the distance to the closest shining bulb. Then, all bulbs go dark again. The goal is to minimize your summed cost. I tell you, Bananistanians are spending their nights playing with bulbs.

Banana day is approaching, and you are hired to play the most beautiful Bulbo game ever. A huge array of bulbs is installed, and you know your initial position and all the light-ups in advance. You need to play the ideal game and impress Bananistanians, and their families.

Input

The first line contains number of turns n and initial position x. Next n lines contain two numbers lstart and lend, which represent that all bulbs from interval [lstart, lend] are shining this turn.

  • 1 ≤ n ≤ 5000
  • 1 ≤ x ≤ 109
  • 1 ≤ lstart ≤ lend ≤ 109

Output

Output should contain a single number which represents the best result (minimum cost) that could be obtained by playing this Bulbo game.

Sample Input

5 4
2 7
9 16
8 10
9 17
1 6
 

Sample Output

8

HINT

题意

有n个区间,你一开始站在x位置

然后n个询问,每个询问给你一个线段,是l,r区间

然后你可以选择走到x位置,花费就是距离

然后关闭这个线段的花费是你现在的位置离线段的最短距离

然后问你依次关闭这n个线段最少需要多少花费

题解:

dp咯,dp[i][j]表示第i轮我在j位置的最小花费

dp[i][j]=min(dp[i][j],dp[i-1][k]+abs(d[i]-d[k]))

然后大概单调栈或者维护一下变成o(n^2)的转移就可以AC了

代码:

//qscqesze
#include <cstdio>
#include <cmath>
#include <cstring>
#include <ctime>
#include <iostream>
#include <algorithm>
#include <set>
#include <bitset>
#include <vector>
#include <sstream>
#include <queue>
#include <typeinfo>
#include <fstream>
#include <map>
#include <stack>
typedef long long ll;
using namespace std;
//freopen("D.in","r",stdin);
//freopen("D.out","w",stdout);
#define sspeed ios_base::sync_with_stdio(0);cin.tie(0)
#define maxn 110
#define eps 1e-9
int Num;
//const int inf=0x7fffffff; //¡ì&szlig;¡ì¨¦¡ì¨¤¡ì¨¦¡§f¡ì3
const int inf=0x3f3f3f3f;
inline ll read()
{
ll x=,f=;char ch=getchar();
while(ch<''||ch>''){if(ch=='-')f=-;ch=getchar();}
while(ch>=''&&ch<=''){x=x*+ch-'';ch=getchar();}
return x*f;
}
//************************************************************************************** struct node
{
ll x,y;
};
node a[];
vector<ll> Q;
ll dp[];
map<ll,int> H;
int main()
{
int n=read();
ll x=read();
Q.push_back(x);
for(int i=;i<=n;i++)
{
a[i].x=read(),a[i].y=read();
Q.push_back(a[i].x);
Q.push_back(a[i].y);
}
sort(Q.begin(),Q.end());
Q.erase(unique(Q.begin(),Q.end()),Q.end());
for(int i=;i<Q.size();i++)
dp[i]=abs(Q[i]-x);
for(int i=;i<Q.size();i++)
H[Q[i]]=i;
for(int i=;i<=n;i++)
{
ll tmp=dp[]-Q[];
for(int j=;j<Q.size();j++)
{
dp[j]=min(dp[j],tmp+Q[j]);
tmp=min(tmp,dp[j]-Q[j]);
}
tmp=dp[Q.size()-]+Q[Q.size()-];
for(int j=Q.size()-;j>=;j--)
{
dp[j]=min(dp[j],tmp-Q[j]);
tmp=min(tmp,dp[j]+Q[j]);
}
for(int j=;j<Q.size();j++)
{
if(Q[j]<a[i].x||Q[j]>a[i].y)dp[j]+=min(abs(Q[j]-a[i].y),abs(Q[j]-a[i].x));
}
/* for(int j=0;j<Q.size();j++)
cout<<dp[j]<<" ";
cout<<endl;*/
}
ll ans = dp[];
for(int i=;i<Q.size();i++)
ans = min(ans,dp[i]);
cout<<ans<<endl;
return ;
}

Codeforces Bubble Cup 8 - Finals [Online Mirror] F. Bulbo DP的更多相关文章

  1. Codeforces Bubble Cup 8 - Finals [Online Mirror] B. Bribes lca

    题目链接: http://codeforces.com/contest/575/problem/B 题解: 把链u,v拆成u,lca(u,v)和v,lca(u,v)(v,lca(u,v)是倒过来的). ...

  2. Codeforces Bubble Cup 8 - Finals [Online Mirror]H. Bots 数学

    H. Bots Time Limit: 1 Sec Memory Limit: 256 MB 题目连接 http://codeforces.com/contest/575/problem/H Desc ...

  3. Codeforces Bubble Cup 8 - Finals [Online Mirror] D. Tablecity 数学题

    D. Tablecity Time Limit: 1 Sec Memory Limit: 256 MB 题目连接 http://codeforces.com/contest/575/problem/D ...

  4. Bubble Cup 12 - Finals Online Mirror, unrated, Div. 1

    Bubble Cup 12 - Finals Online Mirror, unrated, Div. 1 C. Jumping Transformers 我会状压 DP! 用 \(dp[x][y][ ...

  5. Bubble Cup 11 - Finals [Online Mirror, Div. 1]题解 【待补】

    Bubble Cup 11 - Finals [Online Mirror, Div. 1] 一场很好玩的题啊! I. Palindrome Pairs 枚举哪种字符出现奇数次. G. AI robo ...

  6. Bubble Cup X - Finals [Online Mirror]

    来自FallDream的博客,未经允许,请勿转载,谢谢. 组了个菜鸡队打cf上的ACM比赛 比较快做完了8题但是菜的抠脚罚时巨多,所以最后被顶到了19名(居然没出首页) 自己的号自从上次疯狂掉分就没动 ...

  7. Bubble Cup X - Finals [Online Mirror] B. Neural Network country 矩阵快速幂加速转移

    B. Neural Network country time limit per test 2 seconds memory limit per test 256 megabytes Due to t ...

  8. 【简单dfs】Bubble Cup 14 - Finals Online Mirror (Unrated, ICPC Rules, Teams Preferred, Div. 2), problem: (J) Robot Factory,

    传送门  Problem - 1600J - Codeforces 题目   题意 给定n行m列, 求每个连通块由多少格子组成,并将格子数从大到小排序输出 对于每个格子都有一个数(0~15),将其转化 ...

  9. Bubble Cup 12 - Finals [Online Mirror, unrated, Div. 1] E. Product Tuples

    题意略,题解生成函数练习题,1+(q-ai)x卷积即可,线段树优化(类似分治思想) //#pragma GCC optimize(2) //#pragma GCC optimize(3) //#pra ...

随机推荐

  1. Linux常用设置

    1.文件夹操作 创建-->mkdir NAME 删除-->rm NAME -i 删除前逐一询问确认 -f 直接删除,不确认 -r 将目录即以下档案逐一删除 例:删除所有C语言程序文档,删除 ...

  2. Oracle 课程八之性能优化之Oracle SQL Trace

    一. SQL_TRACE 当SQL语句出现性能问题时,我们可以用SQL_TRACE来跟踪SQL的执行情况,通过跟踪,我们可以了解一条SQL或者PL/SQL包的运行情况,SQL_TRACE命令会将SQL ...

  3. Github 终于开始认真考虑开源项目许可证了

    如今GitHub已成为全球最流行的开源项目托管平台,但也有质疑声音——“Github中的大多数项目并不算是开源项目”.这是因为Github中大多数项目并没有明确声明所使用的许可证. 根据版权法规定,如 ...

  4. ASP.NET缓存策略经验谈

    要提升ASP.NET应用程序的性能,最简单.最有效的方式就是使用内建的缓存引擎.虽然也能构建自己的缓存,但由于缓存引擎已提供了如此多的功能,所以完全不必如此麻烦.在很大程度上,ASP.NET开发者在W ...

  5. 状压DP uvalive 6560

    // 状压DP uvalive 6560 // 题意:相邻格子之间可以合并,合并后的格子的值是之前两个格子的乘积,没有合并的为0,求最大价值 // 思路: // dp[i][j]:第i行j状态下的值 ...

  6. Buffer -nodejs

    纯 JavaScript 对 Unicode 友好但是无法很好地处理二进制数据.当我们面对类似 TCP 流或文件系统时,是需要处理八位流的.Node 有几种操作.创建以及消费八位流的策略.原始数据保存 ...

  7. HDU-4747 Mex 线段树

    题目链接:http://acm.hdu.edu.cn/showproblem.php?pid=4747 题意:求一个数列中,所有mex(L,R)的和. 注意到mex是单调不降的,那么首先预处理出mex ...

  8. 【转】Nginx系列(三)--管理进程、多工作进程设计

    原博文出于:http://blog.csdn.net/liutengteng130/article/details/46700999  感谢! Nginx由一个master进程和多个worker进程组 ...

  9. canvas绘制清晰的方法

    很早就开始使用canvas,包括自己绘制各种图形,以及作为画布提供给诸如echarts,当canvas绘制细线条,特别是关于文字绘制会出现很模糊或者锯齿的感觉. <canvas ref=&quo ...

  10. Java设计模式----组合模式(Composit )

    1.  组合模式定义: 组合模式,又叫合成模式,有时又叫部分-整体模式,主要用来描述部分与整体的关系. 定义:将对象组合成树形结构以示" 部分--整体 "的层次结构,使得用户对单个 ...