【CF39E】【博弈论】What Has Dirichlet Got to Do with That?
Description
You all know the Dirichlet principle, the point of which is that if n boxes have no less than n + 1 items, that leads to the existence of a box in which there are at least two items.
Having heard of that principle, but having not mastered the technique of logical thinking, 8 year olds Stas and Masha invented a game. There are a different boxes and b different items, and each turn a player can either add a new box or a new item. The player, after whose turn the number of ways of putting b items into a boxes becomes no less then a certain given number n, loses. All the boxes and items are considered to be different. Boxes may remain empty.
Who loses if both players play optimally and Stas's turn is first?
Input
The only input line has three integers a, b, n (1 ≤ a ≤ 10000, 1 ≤ b ≤ 30, 2 ≤ n ≤ 109) — the initial number of the boxes, the number of the items and the number which constrains the number of ways, respectively. Guaranteed that the initial number of ways is strictly less than n.
Output
Output "Stas" if Masha wins. Output "Masha" if Stas wins. In case of a draw, output "Missing".
Sample Input
2 2 10
Masha
5 5 16808
Masha
3 1 4
Stas
1 4 10
Missing
Hint
In the second example the initial number of ways is equal to 3125.
- If Stas increases the number of boxes, he will lose, as Masha may increase the number of boxes once more during her turn. After that any Stas's move will lead to defeat.
- But if Stas increases the number of items, then any Masha's move will be losing.
Source
/*
酒逢知己千杯少,话不投机半句多。
遥知湖上一樽酒,能忆天涯万里人。
*/ #include <set>
#include <map>
#include <cmath>
#include <cstdio>
#include <vector>
#include <cstring>
#include <cstdlib>
#include <iostream>
#include <algorithm>
#define LOCAL
const int MAXL = ;
const long long MOD = ;
const int MAXK = + ;
const int MAXN = + ;
const int MAXM = ;
using namespace std;
typedef long long LL;
int a , b , n , sg[MAXN][MAXM]; LL pow(int a, int b){
if (b == ) return 1ll;
if (b == ) return (LL)a;
LL tmp = pow(a, b / );
if (b % == ) return tmp * tmp;
else return (LL)tmp * tmp * (LL)a;
}
int dfs (int a , int b) {
if (sg[a][b] >= ) return sg[a][b];
if (pow(a, b) >= (LL)n) {
return sg[a][b] = ;
}
int Ans = ;
Ans |= !dfs(a + , b);//只要有一个是1就能够获胜
Ans |= !dfs(a , b + );
return sg[a][b] = Ans;
}
int work_2(int a){
int b = , turn = ;
int up = (int)sqrt((double)n - 0.0000001);//处理上界
while (a <= up){
if (sg[a][b + ] == ) return (turn == );
a++;
turn = turn ^ ;
}
//剩下的就是判断奇偶性,看谁先到
turn = (turn + (n - - a)) % ;
return turn;
}
int work_1(int b){
int a = ;
if (pow(a, b) >= n) return ;
else {
int turn = ;
for (; !(pow(, b) >= n); b++){
if (sg[a + ][b] == ){
if (turn == ) return ;
else return -;
}
turn ^= ;
}
return ;
}
} int main () { memset(sg , - , sizeof (sg));
scanf("%d%d%d", &a, &b, &n);
if (pow(a, b) >= n) {
puts ("Masha");
return ;
}
dfs(, );
if (a != && b != ){
if (sg[a][b] != ) printf("Masha");
else printf("Stas");
return ;
} if (a == && b == ){
int c = work_1(b + );
int d = work_2(a + );
if (c < || d == ) printf("Masha");
else if (c == ) printf("Missing");
else printf("Stas");
}else if (b == ){
if (work_2(a) != ) printf("Masha");
else printf("Stas");
}else if (a == ){
int tmp = work_1(b);
if (tmp == ) printf("Missing");
else if (tmp == ) printf("Masha");
else printf("Stas");
}
return ;
}
【CF39E】【博弈论】What Has Dirichlet Got to Do with That?的更多相关文章
- CF 39E. What Has Dirichlet Got to Do with That?(记忆化搜索+博弈论)
传送门 解题思路 首先很好写出一个\(O(ab)\)的记搜,但发现这样无法处理\(a=1\)和\(b=1\)的情况,这两种情况需要特判.首先\(a=1\)的情况,就是如果当前选手让\(a+1\)必胜, ...
- IT人生知识分享:博弈论的理性思维
背景: 昨天看了<最强大脑>,由于节目比较有争议性,不知为什么,作为一名感性的人,就想试一下如果自己理性分析会是怎样的呢? 过程是这样的: 中国队(3人)VS英国队(4人). 1:李建东( ...
- LDA( Latent Dirichlet Allocation)主题模型 学习报告
1 问题描述 LDA由Blei, David M..Ng, Andrew Y..Jordan于2003年提出,是一种主题模型,它可以将文档集中每篇文档的主题以概率分布的形式给出,从而通过分析一 ...
- [poj2348]Euclid's Game(博弈论+gcd)
Euclid's Game Time Limit: 1000MS Memory Limit: 65536K Total Submissions: 9033 Accepted: 3695 Des ...
- [综] Latent Dirichlet Allocation(LDA)主题模型算法
多项分布 http://szjc.math168.com/book/ebookdetail.aspx?cateid=1&§ionid=983 二项分布和多项分布 http:// ...
- 博弈论揭示了深度学习的未来(译自:Game Theory Reveals the Future of Deep Learning)
Game Theory Reveals the Future of Deep Learning Carlos E. Perez Deep Learning Patterns, Methodology ...
- TYVJ博弈论
一些比较水的博弈论...(为什么都没有用到那什么SG呢....) TYVJ 1140 飘飘乎居士拯救MM 题解: 歌德巴赫猜想 #include <cmath> #include < ...
- Codeforces 549C. The Game Of Parity[博弈论]
C. The Game Of Parity time limit per test 1 second memory limit per test 256 megabytes input standar ...
- 沃罗诺伊图(Voronoi Diagram,也称作Dirichlet tessellation,狄利克雷镶嵌)
沃罗诺伊图(Voronoi Diagram,也称作Dirichlet tessellation,狄利克雷镶嵌)是由俄国数学家格奥尔吉·沃罗诺伊建立的空间分割算法.灵感来源于笛卡尔用凸域分割空间的思想. ...
随机推荐
- Hadoop的文件读写操作流程
以下主要讲解了Hadoop的文件读写操作流程: 读文件 读文件时内部工作机制参看下图: 客户端通过调用FileSystem对象(对应于HDFS文件系统,调用DistributedFileSystem对 ...
- vijosP1046 观光旅游(最小环)
vijosP1046 观光旅游 链接:https://vijos.org/p/1046 [思路] Floyd求解最小环. [代码] #include<iostream> using nam ...
- Tomcat绑定多个IP地址 多域名绑定
http://blog.csdn.net/stevenyanzhi/article/details/6029776 Tomcat绑定多个IP地址 如果一台服务机上有多个IP地址又有多个工程如何一个IP ...
- v
360导航_新一代安全上网导航 http://www.cnblogs.com/xiaoheimiaoer/p/4309131.html
- poj 1064 Cable master【浮点型二分查找】
Cable master Time Limit: 1000MS Memory Limit: 10000K Total Submissions: 29554 Accepted: 6247 Des ...
- Android 下用 Pull 解析和生成 XML
Java 中是可以用 SAX 和 DOM 解析 XML 的,虽然在 Android 下也可以用这2中方式,但是还是推荐用 Pull.Pull 使用简单,效率相对高,Android 下是集成了 Pul ...
- winform 加密 解密 分类: WinForm 2014-05-16 15:05 400人阅读 评论(0) 收藏
界面显示: 加密: 解密: 代码实现: public string EncryptString(string str) { #region 加密程序 ...
- iOS开发总结-搜索功能实现--使用SKTag
TagsTableViewController.h 文件 #import <UIKit/UIKit.h> #import "personSearch.h" @inter ...
- hdu 2457 DNA repair
AC自动机+DP.按着自动机跑,(其实是生成新的满足题目要求的串,然后找改变最少的.)但是不能跑到是单词的地方,如果跑到单词的话那么说明改变后的串含有病毒了,不满足题意.然后就是应该怎么跑的问题了,现 ...
- wikioi 2573 大顶堆与小顶堆并用
题目描写叙述 Description 我们使用黑匣子的一个简单模型.它能存放一个整数序列和一个特别的变量i.在初始时刻.黑匣子为空且i等于0. 这个黑匣子能运行一系列的命令.有两类命令: ADD(x) ...