Integers, Rationals, and Exponents
·
Mathematics/Real analysis
IntegersDefinition 7.1. The set of integers, denoted $\mathbb{Z}$, is the set $$\{0\} \cup \mathbb{P} \cup -\mathbb{P}$$ where $- \mathbb{P} = \{ -n \, | \, n \in \mathbb{P} \}$. RationalsDefinition 7.2. The set of rational numbers, denoted $\mathbb{Q}$, is the set $$\left\{ \frac{p}{q} \;\middle|\; p, q \in \mathbb{Z} \text{ and } q \neq 0 \right\}$$Integer ExponentsDefinition 7.3. Let $x \in \..
Tensor
·
Physics/수리물리
TensorA tensor of rank $n$ in a $d$-dimensional space란 1부터 $d$까지의 값을 가지는 $n$개의 인덱스로 주어지는 component들을 가지는 object다. 예를 들어 rank 2의 3차원 텐서 $A$는 인덱스가 $2$개이고, 각 인덱스는 1부터 3까지의 값을 갖는다. 따라서 $A_{ij} (1 \leq i, j \leq 3)$과 같이 표현될 수 있고, 이는 3 by 3 matrix의 개념과 동일하다. 이와 같이 기존의 object들을 텐서라는 개념으로 일반화시킬 수 있고, 예컨대 스칼라는 rank 0 텐서, 벡터는 rank 1 텐서, 행렬은 rank 2 텐서이다. Levi-Civita symbol 같은 경우 rank 3 텐서다. Contravariant, ..
Curvilinear Coordinates
·
Physics/수리물리
Curvilinear Coordinates and Scale Factors공간 상의 어떤 점의 좌표가 Cartesian coordinates에서 $(x, y, z)$로 표현된다고 하자. 이때 다른 좌표계에서 이 점의 좌표를 표현했을 때 일반적으로 좌표가 어떻게 얻어지는지 구하고자 한다. 우선 어느 좌표계든 orthogonal하다고 가정한다. 즉 각 좌표계의 단위벡터들은 서로 perpendicular하다. 바꾸고자 하는 좌표계가 $(q_1, q_2, q_3)$로 기술된다고 할 때, $x, y, z$는 각각 $q_1, q_2, q_3$의 함수로 표현된다고 가정하자. $\mathbf{r} = x\mathbf{\hat{x}} + y \mathbf{\hat{y}} + \mathbf{\hat{z}}$라고 할 때,..
Differential Vector Operators
·
Physics/수리물리
GradientCartesian coordinate $\mathbb{R}^3$의 position vector를 $\mathbb{r} = x_1 \mathbf{\hat{e_1}} + x_2 \mathbf{\hat{e_2}} + x_3 \mathbf{\hat{e_3}}$이라 하자. 스칼라 함수 $\varphi : \mathbb{R}^3 \to \mathbb{R}$에 대하여 그 differential은 $$d \varphi = \frac{\partial \varphi}{\partial x_1} dx_1 + \frac{\partial \varphi}{\partial x_2} dx_2 + \frac{\partial \varphi}{\partial x_3} dx_3$$으로 주어진다. 이때 $$\nabla \varp..
Positive Integer
·
Mathematics/Real analysis
Successor Set Definition 6.1. A set $X \subseteq \mathbb{R}$ is said to be a successor set (i) if $1 \in X$,(ii) if $n \in X$, then $n+1 \in X$.실수 집합 $\mathbb{R}$은 그 자체로 successor set이기 때문에,(A9, A1) 위와 같이 정의한 successor set은 적어도 하나는 존재한다고 말할 수 있다. Lemma 6.2Lemma 6.2. If $\mathcal{A}$ is any nonempty collection of successor sets, then $\cap \mathcal{A}$ is a successor set. Proof. Since $1$ is in e..
Real Number
·
Mathematics/Real analysis
Binary OperationDefinition 3.1. A binary operation on a set $X$ is a function from $X \times X$ into $X$. Real Number Definition 3.2. The real numbers $\mathbb{R}$ is a set of objects satisfying Axioms 1 to 13 as listed in the following:(A1) There is a binary operation, called addition and denoted $+$, such that $x, y \in \mathbb{R} \Longrightarrow x + y \in \mathbb{R}$,(A2) $(x + y) + z = x + (..
The Maxwell-Boltzmann Distribution
·
Physics/열물리
Maxwell-Boltzmann DistributionMaxwell-Boltzmann Distribution은 이상 기체를 구성하는 입자의 speed의 확률 분포이다. 즉 이상 기체를 이루는 입자의 속도의 크기, 즉 속력 $v$를 random variable로 두자. 이때 ideal gas는 부피가 0이고 입자간 interaction이 없으므로 완벽하게 random하게 움직인다고 볼 수 있고, 따라서 속도의 방향은 모두 equally probable하다고 가정할 수 있다. 이때 주어진 온도 하에서 속도의 크기의 확률 분포를 알아보자는 것이다. $v$는 continuous하므로 속력이 $v \sim v + dv$라는 구간에서 주어질 확률은 $P(v) dv$로 주어진다. 이때 $x$ 성분만을 고려해서 생각..
The Algorithm of Calculating The Inverse of Triangular Matrices
·
Mathematics/Linear Algebra
주어진 행렬을 LU decomposition을 한 뒤 $L$과 $U$ 행렬 각각의 inverse를 구하면 역행렬을 빠르게 구할 수 있다. 이때 triangular matrix의 inverse를 빠르게 계산하는 방법을 소개하려고 한다. 예컨대 다음과 같은 상삼각 행렬의 역행렬을 계산해 보자. $$U = \begin{pmatrix} 3 & 6 & 8 \\ 0 & 4 & 7 \\ 0 & 0 & 5 \end{pmatrix}$$ 이때 각각의 성분에 대해서 따로따로 생각해 보자. Gauss elimination을 생각하면 역행렬의 대각 성분은 원행렬의 대각 성분의 역수가 된다. $$U^{-1} = \begin{pmatrix} \frac{1}{3} & * & * \\ 0 & \frac{1}{4} & * \\ 0 & ..
A Statistical Definition of Temperature
·
Physics/열물리
Temperature전체 에너지가 $E$로 고정되고 volume과 the nubmer of particles이 고정된 thermodynamic system이 있고, 이 system을 정확히 절반으로 나누어서 각각 $E_1$과 $E_2$의 에너지를 가지고 thermal contact 중에 있어서 열을 교환할 수 있는 independent한 두 개의 계로 구분하자. 이러한 energy distinction의 가정은 충분히 정당화될 수 있다. Thermodynamic limit를 고려할 때 우리는 에너지와 같은 물리량을 extensive variable로 정의할 수 있었다. 즉 system의 size에 dependent한 변수다. 만약 쌍성계와 같이, 중력에 의해 강력하게 interaction하는 system..
Central Limit Theorem
·
Physics/열물리
Central Limit Theorem$N$이 매우 크다고 가정할 때, 서로 독립인 $N$개의 random variable $X_i$에 대해서 $Y = \frac{1}{N}\sum_{i=1}^N X_i$로 정의하자. 각 변수는 동일한 분포를 따르고(어떤 분포인지는 중요하지 않다) fat tail를 가지지 않는, 즉 충분히 빠르게 decay한다고 가정한다. 이런 경우 $Y$의 분포는 정규분포로 근사됨을 보장해주는 정리가 central limit theorem, 중심 극한 정리이다.  위와 같이 정의한 $Y$는 $N$개의 확률 변수 $X_i$들의 산술 평균이다. 각 확률 변수들을 동일한 분포에서 랜덤하게 추출했으므로 평균과 표준편차는 모두 같을 것이고, 이를 각각 $\langle X \rangle$, $\s..