東京大学 情報理工学系研究科 コンピュータ科学専攻 2015年8月実施 専門科目I 問題1
Author
祭音Myyura (co-authored with GPT 5.6 SOL)
Description
Given two -dimensional integer vectors and , let us write if we have for each . Here and denote the -th elements of the vectors and , respectively. Likewise, given two integer matrices and , we write if we have for each . Here and denote the elements in the -th row and the -th column of the matrices and , respectively.
In what follows, a vector all of whose elements are either or is referred to as a “0-1 vector”. A vector each of whose elements is chosen from and , with equal probabilities and independently from the other elements, is referred to as a “random 0-1 vector”. The zero vector (i.e., the vector all of whose elements are ) is denoted by ; and the zero matrix (i.e., the matrix all of whose elements are ) is denoted by .
Answer the following questions.
(1) Let be a random 0-1 vector. Derive the probability with which
holds.
(2) Let be an integer matrix that does not satisfy , and be a random 0-1 vector. Prove that the probability with which holds is no greater than .
(3) Let , and be matrices that do not satisfy , and be a random 0-1 vector. Prove that the probability with which holds is no greater than .
(4) Show an algorithm that: takes three integer matrices , and ; always answers “SATISFIED” if the condition is satisfied; and answers “NOT SATISFIED”, with a probability greater than , if the condition is not satisfied.
题目描述
向量、矩阵之间的等号均按元素模 理解。随机 - 向量是指各分量独立且等概率取 的向量,零向量和零矩阵分别记为 。回答下列问题。
(1)对随机向量 ,求
成立的概率。
(2)设 是不满足 的 整数矩阵, 为随机向量。证明 的概率不超过 。
(3)设 是不满足 的 整数矩阵, 为随机向量。证明 的概率不超过 。
(4)给出一个 随机算法:若 ,总回答“成立”;若 ,则以严格大于 的概率回答“不成立”。
Kai
以下计算均在域 上进行。
(1)
三个方程分别给出 、、,故只有 两个解。因此
(2)
取 的一个非零行,并在该行中取系数为 的位置 。固定除 外的所有分量后,该行与 的内积形如
由于 等概率取 ,该内积为 的条件概率恰为 。事件 还要求其余各行也为 ,故
(3)
令 。由 知 ,且
由(2)立即得到所求概率不超过 。
(4)
独立重复以下检验 次:随机生成 ,依次计算
若某次 ,回答“不成立”;四次均相等时回答“成立”。每次只做三次矩阵向量乘法,故总时间为 。
若 ,算法必定回答“成立”。若 ,由(3),四次均未检出错误的概率至多为 ,所以回答“不成立”的概率至少为