Functional Analysis-Metric Space

Definition

metric space is a pair $(X,d)$ where $X$ is a set and $d$ is a metric(or distance function) $d$ on $X$, that is, a function defined on $X\times X$ such that for all $x,y,z\in X$ we have 4 axioms of a metric:

(1) $d$ is real-valued, finite and nonnegative.

(2) $d(x,y)=0$ if and only if $x=y$

(3) $d(x,y)=d(y,x)$  (Symmetry)

(4) $d(x,y)\leqq d(x,z)+d(z,y)$  (Triangle Inequality)

$X$ is called underlying set of $(X,d)$. If $x,y$ are fixed, we call $d(x,y)$ the distance from $x$ to $y$. A subspace of $(Y,\tilde{d}=d\mid_{ Y\times Y})$ of $(X,d)$ is obtained if we take a subset $Y\subset X$ and restrict $d$ to $Y\times Y$. $\tilde{d}$ is called the metric induced on $Y$ by $d$. Instead of $(X,d)$ we may simply write $X$ if there is no danger of confusion.

Examples

(1)Real line $\textbf{R}$: $(\textbf{R}, d=|\xi-\eta|)$

(2)Euclidean plane: $(\textbf{R}^2, d=\sqrt{(\xi_1-\eta_1)^2+(\xi_2-\eta_2)^2})$

(3)Euclidean plane: $(\textbf{R}^2, d=|\xi_1-\eta_1|+|\xi_2-\eta_2|)$

(4)Unitary space $\textbf{C}^n$: $(\textbf{C}^n, d=\sqrt{|\xi_1-\eta_1|^2+...+|\xi_n-\eta_n|^2})$ where $\textbf{C}^n$ is n dimensional unitary space. Its point $\xi=(\xi_1,...,\xi_n)$ or $\eta=(\eta_1,...,\eta_n)$ is a n dimensional complex number.

(5)Sequence space $l^{\infty}$:  In this space each point is a bounded sequence of complex number. The bounded means that any dimension $\xi_j,~(j=1,2,...)$ of any point $x=(\xi_1,\xi_2,...)$(or briefly $x=(\xi_j)$) in $l^{\infty}$ has a bound $|\xi_j|\leqq c_x$ where $c_x$ depends on point $x$ but doesn‘t depend on $j$. For any 2 points $x$ and $y=(\eta_1,\eta_2,...)$, the metric is defined by $d(x,y)=\mathop{sup}\limits_{j\in N}|\xi_j-\eta_j|,~N\in(1,2,...)$.

(6)Space $l^p$( $p\geq 1$ ): In this space each point is a sequence $x=(\xi_j)=(\xi_1,\xi_2,...),~\xi_j$ can be complex number(complex space) or real number(real space) which must ensure $|\xi_1|^p+|\xi_2|^p+...$ converges, i.e, $\sum^{\infty}_{j=1}|\xi_j|^p \le \infty$. The metric is $d(x,y)=(\sum_{j=1}^{\infty} |\xi_j-\eta_j|^p)^{1/p}$. If $p=2$, we have Hilbert (sequence) space $l^2$.

(7)Sequence space $s$: In this space each point is a bounded or unbounded sequence of complex number, with the metric defined by $d(x,y)=\sum^{\infty}_{j=1}\frac{1}{2^j}\frac{|\xi_j-\eta_j|}{1+|\xi_j-\eta_j|}$.

(8)Function space $C[a,b]$($C$ suggests "continuous"): In this space each point is a function of a real independent variable(自变量) $t$ which is defined and continuous on a given closed interval $J=[a,b]$. The metric is $d(x,y)=\mathop{max} \limits_{t\in J} |x(t)-y(t)|$.

(9)Bounded function space $B(A)$($B$ suggests "bounded"): In this space each point $x\in B(A)$ is a function bounded on a given set $A$, with the metric $d(x,y)=\mathop{sup} \limits_{t\in A} |x(t)-y(t)|$. If $A=[a,b]\subset R$, then we write $B([a,b])$ for $B(A)$.

(10)Discrete metric space: $d(x,x)=0,~~d(x,y)=1~x\neq y$

Verify $l^p$ Space

To demonstrate that a space is a metric space, we must verify the above 4 axioms are all true. Actually, only the last one, i.e, the triangluar inequality, is difficult to verify. Now let‘s take some time to apply the triangular inequality to the example (6),i.e, the $l^p$ space. So our goal is to verify $d(x,y)\leq d(x,z)+d(z,y)$. In the context of $l^p$ space, $d(x,y)=(\sum_{j=1}^{\infty} |\xi_j-\eta_j|^p)^{1/p}$, therefore the goal is to verify $$(\sum_{j=1}^{\infty} |\xi_j-\eta_j|^p)^{1/p} \leq (\sum_{j=1}^{\infty} |\xi_j-\zeta_j|^p)^{1/p} + (\sum_{j=1}^{\infty} |\zeta_j-\eta_j|^p)^{1/p}$$. Before verifying this, let me first illustrate

(A) an auxiliary inequality

(B) the Holder inequality from (A)

(C) the Minkowski inequality from (B)

(D) the triangle inequality (C)

For (A) let $p>1$ and define $q$ by $\frac{1}{p}+\frac{1}{q}=1$ where $p,~q$ are called conjugate exponents. Another form is $1/(p-1)=q-1$, so that

时间: 2024-10-07 16:33:27

Functional Analysis-Metric Space的相关文章

Metric Space

Metric spaces is a large class of spaces on which the closeness of two points is depicted by a distance function, or called a metric. Metric spaces are used to depict the convergence. In the following text, we will construct metrics which are compati

Applied Functional Analysis(Applications to Mathematical Physics ) E.Zeidler

 Applied Functional Analysis(Applications to Mathematical Physics )   E.Zeidler More: QQ565055403

I - Induced Metric Space Gym - 102190I (floyed)

题目链接: I - Induced Metric Space  Gym - 102190I 题目大意:首先是T组测试样例,然后给你一个n*n的矩阵,然后a[i][j]代表i到j的距离是a[i][j].然后-1代表当前的点的距离不知道,让你填空.要求自己到自己距离是0,a[i][j]=a[j][i].a[i][j]为i到j的最短距离. 具体思路:前面是一堆非法情况的判断.然后floyed求最短路,如果当前的点是固定的值并且还可以松弛,那么就是非法的.注意数据范围 floyed竟然打错了.. AC代

RNA_Seq data analysis

原文出处:https://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-0881-8 摘要: RNA-seq已经被广泛的运用,但是现在还没有通行的分析流程.RNA-seq 分析的主要步骤包括:实验设计,QC控制,read alignment,quantification of gene and transcript leves, visualization, differential gene expression, al

Methods for follow-up research of exome analysis:外显子后续分析研究思路总结

外显子后续分析研究思路一般有以下几种(Methods for follow-up research of exome analysis): 1.对突变频率.突变类型.突变方式进行统计分析 Mutations statistical analysis 具体见下图: 参考文献:Di, Jiabo, et al. "Whole exome sequencing reveals intertumor heterogeneity and distinct genetic origins of sporad

[转]林达华推荐的几本数学书

http://blog.csdn.net/lqhbupt/article/details/32106217 Dahua Lin早在几年前就已经冒尖出来了,现在在MIT攻读博士学位,前途不可限量.他总是有无穷的精力,学习,同时几篇几篇的写paper,几万行几万行的写code,几万字几万字的写blog.他扎实的数学功底和相关知识的功底,以及深睿的洞察和理解问题的能力,注定他将在machine learning和computer vision等相关领域取得大量的成果,甚至是突破性的成果.期待他在这些领

和机器学习和计算机视觉相关的数学(转载)

1. 线性代数 (Linear Algebra): 我想国内的大学生都会学过这门课程,但是,未必每一位老师都能贯彻它的精要.这门学科对于Learning是必备的基础,对它的透彻掌握是必不可少的.我在科大一年级的时候就学习了这门课,后来到了香港后,又重新把线性代数读了一遍,所读的是Introduction to Linear Algebra (3rd Ed.)  by Gilbert Strang.这本书是MIT的线性代数课使用的教材,也是被很多其它大学选用的经典教材.它的难度适中,讲解清晰,重要

【机器学习系列】机器学习界大牛林达华推荐的书籍

Recommended Books Here is a list of books which I have read and feel it is worth recommending to friends who are interested in computer science. Machine Learning Pattern Recognition and Machine Learning Christopher M. Bishop A new treatment of classi

[转]和机器学习和计算机视觉相关的数学

http://blog.sina.com.cn/s/blog_6833a4df0100nazk.html 1. 线性代数 (Linear Algebra): 我想国内的大学生都会学过这门课程,但是,未必每一位老师都能贯彻它的精要.这门学科对于Learning是必备的基础,对它的透彻掌握是必不可少的.我在科大一年级的时候就学习了这门课,后来到了香港后,又重新把线性代数读了一遍,所读的是 Introduction to Linear Algebra (3rd Ed.) by Gilbert Stra