Lecture 1: Rn\mathbb{R}^n and Cn\mathbb{C}^n; Definition of Vector Space (域 与 向量空间)

30 分钟阅读

1. Introduction

线性代数是研究有限维向量空间上的线性映射的学科。

课程路线:

C⟶Lists⟶Fn⟶Vector Space⟶Subspaces\mathbb{C} \longrightarrow \text{Lists} \longrightarrow F^n \longrightarrow \text{Vector Space} \longrightarrow \text{Subspaces}
  • 在实数与复数同时研究时,能得到更好的定理和更深的洞察。
  • 将平面(R2\mathbb{R}^2)和普通空间(R3\mathbb{R}^3)推广到 Rn\mathbb{R}^n 和 Cn\mathbb{C}^n。
  • 再进一步抽象为向量空间(vector space)的一般概念。
  • 子空间(subspace)之于向量空间,类似于子集之于集合。

2. Complex Numbers

2.1 Definition

Definition 1.1 (Complex Numbers, C\mathbb{C})

  • 一个复数是一个有序对 (a,b)(a,b),其中 a,b∈Ra, b \in \mathbb{R},记作 a+bia + bi。
  • 所有复数的集合记为 C={a+bi:a,b∈R}\mathbb{C} = \{a + bi : a, b \in \mathbb{R}\}。
  • 加法和乘法定义为: (a+bi)+(c+di)=(a+c)+(b+d)i(a+bi) + (c+di) = (a+c) + (b+d)i (a+bi)(c+di)=(ac−bd)+(ad+bc)i(a+bi)(c+di) = (ac - bd) + (ad + bc)i
  • 将 a+0ia + 0i 等同于实数 aa,故 R⊂C\mathbb{R} \subset \mathbb{C}。
  • 符号 ii 由 Euler 于 1777 年首次使用,表示 −1\sqrt{-1}。
  • 乘法公式的动机:假设 i2=−1i^2 = -1,利用通常的算术规则展开即可。

2.2 Properties of Complex Arithmetic (1.3)

性质内容
交换律α+β=β+α\alpha + \beta = \beta + \alpha,αβ=βα\alpha\beta = \beta\alpha
结合律(α+β)+λ=α+(β+λ)(\alpha+\beta)+\lambda = \alpha+(\beta+\lambda),(αβ)λ=α(βλ)(\alpha\beta)\lambda = \alpha(\beta\lambda)
单位元λ+0=λ\lambda + 0 = \lambda,λ⋅1=λ\lambda \cdot 1 = \lambda
加法逆元对每个 α∈C\alpha \in \mathbb{C},存在唯一 β\beta 使 α+β=0\alpha + \beta = 0
乘法逆元对每个 α≠0\alpha \neq 0,存在唯一 β\beta 使 αβ=1\alpha\beta = 1
分配律λ(α+β)=λα+λβ\lambda(\alpha + \beta) = \lambda\alpha + \lambda\beta

2.3 Notation: −α-\alpha, Subtraction, 1/α1/\alpha, Division

  • −α-\alpha:α\alpha 的加法逆元
  • β−α=β+(−α)\beta - \alpha = \beta + (-\alpha)
  • 1/α1/\alpha:α≠0\alpha \neq 0 时的乘法逆元
  • β/α=β(1/α)\beta / \alpha = \beta(1/\alpha)

3. Notation: FF

Notation 1.6

本书中 FF 表示 R\mathbb{R} 或 C\mathbb{C}。


4. Lists

Definition 1.8 (List, Length)

  • 长度为 nn 的列表(list) 是 nn 个元素的有序集合。
  • 两个列表相等   ⟺  \iff 长度相同且对应位置的元素相同。

列表与集合的区别:

  • 列表中顺序有意义:(3,5)≠(5,3)(3,5) \neq (5,3),但 {3,5}={5,3}\{3,5\} = \{5,3\}。
  • 列表中重复有意义:(4,4)≠(4,4,4)(4,4) \neq (4,4,4),但 {4,4}={4,4,4}={4}\{4,4\} = \{4,4,4\} = \{4\}。

许多数学家将长度为 nn 的列表称为 nn-元组(nn-tuple)。


5. FnF^n

Definition 1.11 (FnF^n, Coordinate)

Fn={(x1,…,xn):xk∈F for k=1,…,n}F^n = \{(x_1, \dots, x_n) : x_k \in F \text{ for } k = 1, \dots, n\}

其中 xkx_k 称为第 kk 个坐标(coordinate)。

5.1 运算定义

运算定义
加法(x1,…,xn)+(y1,…,yn)=(x1+y1,…,xn+yn)(x_1,\dots,x_n) + (y_1,\dots,y_n) = (x_1+y_1,\dots,x_n+y_n)
零元0=(0,…,0)0 = (0, \dots, 0)
加法逆元−x=(−x1,…,−xn)-x = (-x_1, \dots, -x_n),满足 x+(−x)=0x + (-x) = 0
标量乘法λ(x1,…,xn)=(λx1,…,λxn)\lambda(x_1,\dots,x_n) = (\lambda x_1, \dots, \lambda x_n)

5.2 基本性质

  • 加法交换律(1.14):x+y=y+xx + y = y + x
  • 注意:符号 00 在不同语境下含义不同——可以是 FnF^n 中的零向量,也可以是 FF 中的数 00。

6. Fields(域)

Definition 一个域(field) 是一个包含至少两个不同元素 00 和 11 的集合,配备加法和乘法运算,满足 1.3 中列出的所有性质(交换律、结合律、单位元、逆元、分配律)。

例子:

  • R\mathbb{R}(实数)
  • C\mathbb{C}(复数)
  • Q\mathbb{Q}(有理数)
  • {0,1}\{0, 1\},其中 1+1=01 + 1 = 0(即 Z2\mathbb{Z}_2)

注意: 本书中只处理 R\mathbb{R} 和 C\mathbb{C},但许多定义和定理对任意域同样成立。


7. Vector Space(向量空间)

7.1 动机

从 FnF^n 的运算性质中抽象出来:加法交换、结合、有单位元、有逆元;标量乘法结合;分配律连接两种运算。

7.2 Definition

下面用 定义 1.20 把这套运算公理正式固定下来。

Definition 1.20 (Vector Space)

一个向量空间是一个集合 VV,配备加法 +:V×V→V+: V \times V \to V 和标量乘法 ⋅:F×V→V\cdot: F \times V \to V,满足以下六条公理:

编号公理内容
(1)交换律u+v=v+uu + v = v + u,∀ u,v∈V\forall\, u,v \in V
(2)结合律(u+v)+w=u+(v+w)(u+v)+w = u+(v+w) 且 (ab)v=a(bv)(ab)v = a(bv)
(3)加法单位元∃ 0∈V\exists\, 0 \in V 使得 v+0=vv + 0 = v,∀ v∈V\forall\, v \in V
(4)加法逆元∀ v∈V\forall\, v \in V,∃ w∈V\exists\, w \in V 使得 v+w=0v + w = 0
(5)乘法单位元1v=v1v = v,∀ v∈V\forall\, v \in V
(6)分配律a(u+v)=au+ava(u+v) = au + av 且 (a+b)v=av+bv(a+b)v = av + bv

7.3 相关术语

  • VV 的元素称为向量(vector)或点(point)。
  • 若 F=RF = \mathbb{R},称 VV 为实向量空间;若 F=CF = \mathbb{C},称 VV 为复向量空间。
  • 最简单的向量空间是 {0}\{0\}。

7.4 例子

  • FnF^n(通常的加法和标量乘法)
  • F∞F^\infty:所有 FF 中元素构成的序列
  • FSF^S:从集合 SS 到 FF 的所有函数,加法和标量乘法逐点定义: (f+g)(x)=f(x)+g(x),(λf)(x)=λf(x)(f+g)(x) = f(x) + g(x), \quad (\lambda f)(x) = \lambda f(x) 加法单位元是零函数 0(x)=00(x) = 0,ff 的加法逆元是 −f-f。

8. Elementary Properties

8.1 唯一性

Theorem 1.26 (Unique Additive Identity)

向量空间的加法单位元唯一。

证明思路: 设 00 和 0′0' 都是加法单位元,则 0′=0′+0=0+0′=00' = 0' + 0 = 0 + 0' = 0。

Theorem 1.27 (Unique Additive Inverse)

每个元素的加法逆元唯一。

证明思路: 设 ww 和 w′w' 都是 vv 的加法逆元,则 w=w+0=w+(v+w′)=(w+v)+w′=0+w′=w′w = w + 0 = w + (v + w') = (w+v) + w' = 0 + w' = w'。

8.2 基本等式

结果内容证明关键
1.300v=00v = 0(标量 00 乘任意向量 = 零向量)分配律:0v=(0+0)v=0v+0v0v = (0+0)v = 0v + 0v
1.31a0=0a0 = 0(任意标量乘零向量 = 零向量)分配律:a0=a(0+0)=a0+a0a0 = a(0+0) = a0 + a0
1.32(−1)v=−v(-1)v = -vv+(−1)v=1v+(−1)v=(1+(−1))v=0v=0v + (-1)v = 1v + (-1)v = (1+(-1))v = 0v = 0

8.3 消去律(Cancellation Law)

Theorem (Cancellation Law)

若 x+z=y+zx + z = y + z,则 x=yx = y。

证明: 取 vv 使得 z+v=0z + v = 0,则

x=x+0=x+(z+v)=(x+z)+v=(y+z)+v=y+(z+v)=y+0=y.■x = x + 0 = x + (z+v) = (x+z) + v = (y+z) + v = y + (z+v) = y + 0 = y. \quad \blacksquare

9. 反例:验证公理的必要性

Example. 设 V={(a1,a2):a1,a2∈R}V = \{(a_1, a_2) : a_1, a_2 \in \mathbb{R}\},定义

(a1,a2)+(b1,b2)=(a1+2b1,  a2+2b2)(a_1, a_2) + (b_1, b_2) = (a_1 + 2b_1,\; a_2 + 2b_2) c(a1,a2)=(ca1,ca2)c(a_1, a_2) = (ca_1, ca_2)

问:VV 在此运算下是否为 R\mathbb{R} 上的向量空间?

答:不是。 加法不满足交换律:

(a1,a2)+(b1,b2)=(a1+2b1,a2+2b2)(a_1, a_2) + (b_1, b_2) = (a_1 + 2b_1, a_2 + 2b_2) (b1,b2)+(a1,a2)=(b1+2a1,b2+2a2)(b_1, b_2) + (a_1, a_2) = (b_1 + 2a_1, b_2 + 2a_2)

一般地 a1+2b1≠b1+2a1a_1 + 2b_1 \neq b_1 + 2a_1,故公理 (1) 失败。


10. 概念联系图

ℂ (复数:具体代数结构)
 │
 ▼
Lists (有序元组:强调顺序与重复)
 │
 ▼
Fⁿ (坐标向量:加法 + 标量乘法)
 │  抽象出运算公理
 ▼
Vector Space V over Field F
 │  (下节课)
 ▼
Subspace (子空间)

核心思想: 每一步都是”去掉具体结构,保留运算公理”的抽象化过程。FnF^n 是向量空间的原型,向量空间定义中的六条公理恰好是 FnF^n 运算满足的所有代数性质。


11. Exercises (from LADR, Sheldon Axler)

复数运算(1.A)

题号考查内容
1.A.1证明 α+β=β+α\alpha + \beta = \beta + \alpha,∀ α,β∈C\forall\, \alpha, \beta \in \mathbb{C}(加法交换律)
1.A.3证明 (αβ)λ=α(βλ)(\alpha\beta)\lambda = \alpha(\beta\lambda),∀ α,β,λ∈C\forall\, \alpha, \beta, \lambda \in \mathbb{C}(乘法结合律)

FnF^n 的运算验证(1.A)

题号考查内容
1.A.11证明 (x+y)+z=x+(y+z)(x+y)+z = x+(y+z),∀ x,y,z∈Fn\forall\, x,y,z \in F^n(加法结合律)
1.A.12证明 (ab)x=a(bx)(ab)x = a(bx),∀ x∈Fn\forall\, x \in F^n,a,b∈Fa,b \in F(标量乘法结合律)
1.A.14证明 λ(x+y)=λx+λy\lambda(x+y) = \lambda x + \lambda y,∀ λ∈F\forall\, \lambda \in F,x,y∈Fnx,y \in F^n(分配律)

向量空间公理与抽象证明(1.B)

题号考查内容
1.B.2若 a∈Fa \in F,v∈Vv \in V,且 av=0av = 0,证明 a=0a = 0 或 v=0v = 0
1.B.3若 v,w∈Vv, w \in V,解释为何存在唯一 x∈Vx \in V 使得 v+3x=wv + 3x = w
1.B.4空集不是向量空间——它恰好违反了哪一条公理?(答:加法单位元的存在性)
1.B.5证明在向量空间定义中,“加法逆元”条件可替换为”0v=00v = 0,∀v∈V\forall v \in V”

1.A.1 Show that α+β=β+α\alpha + \beta = \beta + \alpha for all α,β∈C\alpha,\beta \in \mathbb{C}.

sol:
Axiom: ∀ m,n∈R, m+n=n+m\forall\ m,n \in \mathbb{R},\ m + n = n + m.
Let α=a+bi,β=c+di\alpha = a+bi,\beta=c+di, where a,b,c,d∈Ra,b,c,d \in \mathbb{R},
then α+β=a+bi+c+di:=(a+c)+(b+d)i\alpha + \beta = a + bi + c + di := (a + c) + (b + d)i ,
and β+α=c+di+a+bi:=(c+a)+(d+b)i\beta + \alpha = c + di + a + bi := (c + a) + (d + b)i.
By applying the axiom, α+β=(a+c)+(b+d)i=(c+a)+(d+b)i=β+α\alpha + \beta = (a + c) + (b + d)i = (c + a) + (d + b)i = \beta + \alpha.

1.A.3 Show that (αβ)λ=α(βλ)(\alpha \beta)\lambda = \alpha(\beta\lambda) for all α,β,λ∈C\alpha,\beta,\lambda \in \mathbb{C}.

sol:
Let α=a+bi,β=c+di,λ=e+fi\alpha = a + bi, \beta = c + di, \lambda = e + fi, where a,b,c,d,e,f∈Ra,b,c,d,e,f \in \mathbb{R},
then

(αβ)λ=((a+bi)(c+di))(e+fi)=((𝑎𝑐−𝑏𝑑)+(𝑎𝑑+𝑏𝑐)𝑖)(e+fi)(\alpha\beta)\lambda = ((a + bi)(c + di))(e + fi) = ((𝑎𝑐 −𝑏𝑑) +(𝑎𝑑 +𝑏𝑐)𝑖)(e + fi)

by definition of multiplication on C\mathbb{C}.
In next step, we apply the definition again on (𝑎𝑐−𝑏𝑑)+(𝑎𝑑+𝑏𝑐)𝑖(𝑎𝑐 −𝑏𝑑) +(𝑎𝑑 +𝑏𝑐)𝑖 and e+fie + fi.
Thus, we gain (αβ)λ=(ace−bde−adf−bcf)+(ade+bce+acf−bdf)i(\alpha\beta)\lambda = (ace - bde - adf - bcf) + (ade + bce + acf - bdf)i.
Similarly,

α(βλ)=(a+bi)((ce−df)+(de+cf)i)=(ace−adf−bde−bcf)+(ade+acf+bce−bdf)i=(ace−bde−adf−bcf)+(ade+bce+acf−bdf)i=(αβ)λ\begin{aligned} \alpha(\beta\lambda) &= (a + bi)((ce - df) + (de + cf)i) \\ &= (ace - adf - bde - bcf) + (ade + acf + bce - bdf)i\\ &= (ace - bde - adf - bcf) + (ade + bce + acf - bdf)i\\ &= (\alpha\beta)\lambda \end{aligned}

1.A.11 Show that (x+y)+z=x+(y+z)(x + y)+z = x+(y+z) for all x,y,z∈Fnx,y,z ∈ \mathbb{F}^n

sol:
Let:

x=(x1,x2,...,xn),y=(y1,y2,...,yn),z=(z1,z2,...,zn).\begin{aligned} &x = (x_1,x_2,...,x_n),\\ &y = (y_1,y_2,...,y_n),\\ &z = (z_1,z_2,...,z_n). \end{aligned}

(x+y)+z=(x1+y1,x2+y2,...,xn+yn)+z (x+y)+z=(x_1+y_1,x_2+y_2,...,x_n+y_n)+z\ by definition of addition in Fn\mathbb{F}^n.
Denotes (x1+y1,x2+y2,...,xn+yn)(x_1+y_1,x_2+y_2,...,x_n+y_n) as a vector γ∈Fn\gamma \in \mathbb{F}^n,
then

(x+y)+z=γ+z=((x1+y1)+z1,(x2+y2)+z2,...,(xn+yn)+zn)=(x1+y1+z1,x2+y2+z2,...,xn+yn+zn)(associativity in R)\begin{aligned} (x + y) + z &= \gamma + z \\&= ((x_1+y_1)+z_1,(x_2+y_2)+z_2,...,(x_n+y_n)+z_n) \\&= (x_1+y_1+z_1,x_2+y_2+z_2,...,x_n+y_n+z_n)\quad (\text{associativity in}\ \mathbb{R}) \end{aligned}

Similarly,

x+(y+z)=x+(y1+z1,y2+z2,...,yn+zn)=(x1+(y1+z1),x2+(y2+z2),...,xn+(yn+zn))=(x1+y1+z1,x2+y2+z2,...,xn+yn+zn)(associativity in R)=(x+y)+z\begin{aligned} x + (y + z) &= x + (y_1+z_1,y_2+z_2,...,y_n+z_n) \\ &= (x_1+(y_1+z_1),x_2+(y_2+z_2),...,x_n+(y_n+z_n))\\ &= (x_1+y_1+z_1,x_2+y_2+z_2,...,x_n+y_n+z_n)\quad (\text{associativity in}\ \mathbb{R})\\ &= (x + y) + z \end{aligned}

1.A.12 Show that (ab)x=a(bx)(ab)x = a(bx) for all 𝑥∈Fn𝑥 ∈ \mathbb{F}^n and all a,b∈Fa,b ∈ \mathbb{F}

sol:
Let xx be (x1,x2,...,xn)∈Fn and a,b∈F(x_1,x_2,...,x_n) \in \mathbb{F}^n\ \text{and}\ a,b \in \mathbb{F} .
By the definition of scalar multiplication in Fn\mathbb{F}^n,

(ab)x=((ab)x1,(ab)x2,...,(ab)xn)=(abx1,abx2,...,abxn),\begin{aligned} (ab)x &= ((ab)x_1,(ab)x_2,...,(ab)x_n) \\&= (abx_1,abx_2,...,abx_n), \end{aligned}

and

(bx)=(bx1,bx2,...,bxn):=y∈Fn.(bx) = (bx_1,bx_2,...,bx_n) := y \in \mathbb{F}^n.

Then by the associativity of multiplication in the field F\mathbb{F}, we know that (ab)xi=a(bxi)(ab)x_i = a(bx_i) for each i=1,2,...,ni = 1,2,...,n. Thus,

a(bx)=ay=(ay1,ay2,...,ayn)=(a(bx1),a(bx2),...,a(bxn))=(abx1,abx2,...,abxn)=(ab)x.\begin{aligned} a(bx) &= ay\\ &= (ay_1,ay_2,...,ay_n)\\ &= (a(bx_1),a(bx_2),...,a(bx_n)) \\&= (abx_1,abx_2,...,abx_n) \\&= (ab)x. \end{aligned}

1.A.14 Show that λ(x+y)=λx+λy\lambda(x + y) = \lambda x + \lambda y for all λ∈F\lambda ∈ \mathbb{F} and all x,y∈Fnx,y ∈ \mathbb{F}^n.

sol:
By the definition of addition in Fn\mathbb{F}^n,

x+y=(x1,...,xn)+(y1,...,yn)=(x1+y1,...,xn+yn).x + y = (x_1, ...,x_n) + (y_1,...,y_n) = (x_1 +y_1,...,x_n + y_n).

And by the definition of scalar multiplication in Fn\mathbb{F}^n,

λ(x+y)=(λ(x1+y1),...,λ(xn+yn))=(λx1+λy1,...,λxn+λyn)=(λx1,...,λxn)+(λy1,...,λyn)=λx+λy\begin{aligned} \lambda(x + y) = (\lambda(x_1 + y_1), ..., \lambda(x_n + y_n)) &= (\lambda x_1 + \lambda y_1, ..., \lambda x_n + \lambda y_n)\\ &= (\lambda x_1, ..., \lambda x_n) + (\lambda y_1, ..., \lambda y_n)\\ &= \lambda x + \lambda y \end{aligned}

1.B.2 Suppose a∈F,v∈Va ∈ \mathbb{F},v ∈ V,and av=0av = 0. Prove that a=0a = 0 or v=0v = 0.

sol:
Suppose a≠0a \neq 0:
Since a∈Fa ∈ \mathbb{F}, ∃ a−1∈F s.t. a−1⋅a=1\exists\ a^{-1} \in \mathbb{F}\ s.t.\ a^{-1} \cdot a = 1.
By the associativity of scalar multiplication, a−1(av)=(a−1a)va^{−1} (av)=(a^{−1} a)v
Then multiply a−1a^{-1} on both side of equation, \

a−1⋅av=a−1⋅0 ⇒associativity of scalar multiplication 1⋅v=0 ⇒multiplicative identity v=0a^{-1} \cdot av = a^{-1} \cdot 0 \ \overset{\text{associativity of scalar multiplication}}\Rightarrow\ 1 \cdot v = 0\ \overset{\text{multiplicative identity}}\Rightarrow\ v = 0

Conversely, if we suppose a=0a = 0: The equality obviously holds.
Thus, we have shown that if a≠0a \neq 0, v=0v = 0. This is logically equivalent to saying that a=0a = 0 or v=0v = 0.

1.B.3 Suppose v,w∈Vv,w ∈ V. Explain why there exists a unique x∈Vx ∈ V such that v+3x=wv + 3x = w.

sol:
Existence:
Since v∈Vv \in V, ∃ −v∈V s.t. v+(−v)=0\exists\ -v \in V\ s.t.\ v + (-v) = 0.
Let x=13(w+(−v))x = \frac{1}{3}(w+(− v)). Then

v+3x=v+3(13(w+(−v)))=v+(3⋅13)(w+(−v))(associativity of scalar multiplication)=v+1(w+(−v))(field arithmetic)=v+(w+(−v))(multiplicative identity)=v+((−v)+w)(commutativity of addition)=(v+(−v))+w(associativity of addition)=0+w(additive inverse)=w(additive identity)\begin{aligned} v + 3x &= v + 3\left(\frac{1}{3}(w + (-v))\right) \\ &= v + \left(3 \cdot \frac{1}{3}\right)(w + (-v)) \quad \text{(associativity of scalar multiplication)} \\ &= v + 1(w + (-v)) \quad \text{(field arithmetic)} \\ &= v + (w + (-v)) \quad \text{(multiplicative identity)} \\ &= v + ((-v) + w) \quad \text{(commutativity of addition)} \\ &= (v + (-v)) + w \quad \text{(associativity of addition)} \\ &= 0 + w \quad \text{(additive inverse)} \\ &= w \quad \text{(additive identity)} \end{aligned}

Thus, such an xx exists.

Uniqueness:
Suppose x1,x2∈Vx_1, x_2 \in V both satisfy the equation, meaning:
v+3x1=wandv+3x2=wv + 3x_1 = w \quad \text{and} \quad v + 3x_2 = w
Therefore, v+3x1=v+3x2.v + 3x_1 = v + 3x_2.
Add −v-v to both the side (same process as proving the Cancellation Law for Vector Addition):

−v+v+3x1=−v+v+3x2⇒0+3x1=0+3x2⇒3x1=3x2⇒13⋅3x1=13⋅3x2⇒1⋅x1=1⋅x2⇒Multiplicative Identity x1=x2\begin{aligned} -v + v + 3x_1 &= -v + v + 3x_2\\ \Rightarrow 0 + 3x_1 &= 0 + 3x_2\\ \Rightarrow 3x_1 &= 3x_2\\ \Rightarrow \frac{1}{3} \cdot 3x_1 &= \frac{1}{3} \cdot 3x_2\\ \Rightarrow 1 \cdot x_1 &= 1 \cdot x_2\\ \overset{\text{Multiplicative Identity}}\Rightarrow\ x_1 &= x_2 \end{aligned}

Thus, the solution is unique.

1.B.4 The empty set is not a vector space. The empty set fails to satisfy only one of the requirements listed in the definition of a vector space (1.20). Which one?

sol:
The additive identity because there isn’t a definition of 0.

This is an existential statement ( ∃ 0 ∈ V , … ).
Since the empty set contains no elements at all, no such element 0 can exist.
Therefore this axiom is false for V=∅V = ∅.

All other axioms in Definition 1.20 — including:
commutativity, associativity, additive inverse, multiplicative identity, and the distributive properties — are universal statements of the form “for all v ∈ V , …” or “for all u , v ∈ V , … ”.

When V=∅V = ∅ , these statements are vacuously true, because there are no elements in VV that could serve as counterexamples.
In particular, the additive inverse axiom (“for every v∈Vv ∈ V , there exists w∈Vw ∈ V such that v+w=0v + w = 0”) is vacuously true:
since there is no v∈∅v ∈ ∅ to check, the condition is satisfied by default. Thus, the additive identity is the only axiom that fails for the empty set.

1.B.5 Show that in the definition of a vector space (1.20), the additive inverse condition can be replaced with the condition that

0v=0 for all v∈V0v = 0\ \text{for all}\ v \in V

Here the 0 on the left side is the number 0, and the 0 on the right side is the additive identity of VV. The phrase a “condition can be replaced” in a definition means that the collection of objects satisfying the definition is unchanged if the original condition is replaced with the new condition.

sol:
Part I: 0v=0 for all v∈V ⇒ ∀ v∈V, ∃ w∈V s.t. v+w=00v = 0\ \text{for all}\ v \in V\ \Rightarrow\ \forall\ v \in V,\ \exists\ w \in V\ s.t.\ v+w=0:
Let vv be any of the vector in VV.
Then (−1)v(-1)v is exactly the additive inverse of vv. Following is the proof:\

v+(−1)v=1v+(−1)v(multiplicative identity)=(1+(−1))v(distributive property)=0v=0(0v=0 for all v∈V)\begin{aligned} v + (-1)v &= 1v + (-1)v \quad \text{(multiplicative identity)}\\ &= (1+(-1))v \quad \text{(distributive property)}\\ &= 0v\\ &= 0 \quad (0v = 0\ \text{for all}\ v \in V) \end{aligned}

Part II: 0v=0 for all v∈V ⇐ ∀ v∈V, ∃ w∈V s.t. v+w=00v = 0\ \text{for all}\ v \in V\ \Leftarrow\ \forall\ v \in V,\ \exists\ w \in V\ s.t.\ v+w=0:
Let ww be the additive inverse of 0v0v:

0v=(0+0)v0v=0v+0v(distributive property)0v+w=0v+0v+w0v+w=0v+(0v+w)(associativity)0=0v+0(additive inverse)0=0v(additive identity)\begin{aligned} 0v &= (0+0)v\\ 0v &= 0v+0v \quad \text{(distributive property)}\\ 0v + w &= 0v + 0v + w\\ 0v + w &= 0v + (0v + w) \quad \text{(associativity)}\\ 0 &= 0v + 0 \quad \text{(additive inverse)}\\ 0 &= 0v \quad \text{(additive identity)} \end{aligned}

Thus 0v=0 for all v∈V ⇔ ∀ v∈V, ∃ w∈V s.t. v+w=00v = 0\ \text{for all}\ v \in V\ \Leftrightarrow\ \forall\ v \in V,\ \exists\ w \in V\ s.t.\ v+w=0


12. Questions to Carry Forward

  • 如何判断一个子集是否为子空间?(下节课:三个条件)
  • 子空间的和(sum)与直和(direct sum)如何定义?
  • 为什么”更多变量齐次方程必有非零解”可以用向量空间语言证明?