大學數學
數學 大 學 線性代數 , 大
數 .
大 學 線性代數學 ,
學 線性代數 . 大學線性代數
. 學 大 大 ( ) 線性
代數 , . 大學 線性代數 , 大
學 數學 . 大 大 數學
, 學 , 線性代數
. , 線性代數
數學 . , .
, .
學 , 數學 學 , 線性
代數 . 線性代數 .
, . 學
. , (Question).
, 大
. , 線性代數 . ,
學 線性代數 學 線性代數 , .
, ,
代. , .
, . , 性
, . , .
v
Linear Transformation of General Vector
Space
, 前 vector space. vector space
vector space linear transformation. 前 , vector space
性 , Rn . 前 .
6.1. Vector Space and Subspace
Rm , 線性 Rm ,
Proposition 1.2.3 8 , vector space.
vector space vector space subspaces 性 .
V , V +, V u, v∈ V,
u + v V ( 性). R V
數 r∈ R v∈ V, r v rv V ( 數
性).
Definition 6.1.1. V +, R V 數 .
8 性 , V vector space.
(1) u, v∈ V, u + v = v + u.
(2) u, v, w∈ V, (u + v) + w = u + (v + w).
(3) O∈ V u∈ V O + u = u.
(4) u∈ V u′∈ V u + u′= O.
(5) r, s∈ R u∈ V, r(su) = (rs)u.
(6) r, s∈ R u∈ V, (r + s)u = ru + su.
111
(7) r∈ R u, v∈ V r(u + v) = ru + rv.
(8) u∈ V, 1u = u.
, vector space,
數 . vector space , V vector
space + , 數 數 ,
數 . vector space, Rn,
數 , 數 .
vector space , 數 . vector space
數 數 , “field” 數 . over
field vector space ( over field 大). 學
field 數 R ,
overR vector space, vector space .
Corollary 1.2.4 Rn , 性 (3) O (4)
u, u′ . Rn 數 ,
8 性 , vector space .
Proposition 6.1.2. V vector space, V O
u∈ V O + u = u. , u∈ V, u′∈ V u + u′= O.
O , O V O + u = u,∀u ∈ V
, V additive identity zero element. u∈ V,
u′ u + u′= O, −u u′,
u additive inverse.
, 性 (3) O u∈ V
O + u = u ( Proposition 6.1.2 u ),
u∈ V w + u = u, w = O. Proposition 6.1.2
, u∈ V, w + u = u, w = O . 性
O 性 . 數 r, rO = O.
性 (7) rO + rO = r(O + O) = rO. rO rO rO, 前
rO = O. , , Corollary 1.2.6
Corollary 1.2.7.
Proposition 6.1.3. V vector space, .
(1) w∈ V u∈ V w + u = u, w = O.
(2) v∈ V 0v = O.
(3) r∈ R rO = O.
(4) r∈ R,v ∈ V (−1)(rv) = −(rv) = r(−v).
Proposition 6.1.3 r = 0 v = O rv = O, r̸= 0 v̸= O,
rv = O ? . r̸= 0 v̸= O, 1/r rv
vector space 性 (5) 1r(rv) =rrv =1v, 性 (8) 1v = v, 1r(rv) = v̸= O.
rv = O, Proposition 6.1.3 (3) 1r(rv) = 1rO = O, . rv O.
Question 6.1. V vector space, v∈ V v̸= O, r, s∈ R r̸= s,
rv̸= sv, V vector space over R V , V
.
言 , vector space 數 8 性 , vector
space 數 . , 數 “ ” ,
w + (−v) w− v. 大
. 2u + v = w, 1/2 u = 12(w− v).
vector space .
Example 6.1.4. (A) Mm×n m× n matrices .
數 ( Definition 3.1.2), vector space
8 性 ( Proposition 3.1.3). 數 Mm×n
vector space.
(B) P(R) x 數 數 . 數 ,
P(R) vector space. f (x) = anxn+··· + a1x + a0, g(x) = bmxm+··· + b1x + b0∈ P(R), m≤ n, g(x) g(x) = bnxn+···+bmxm+···+b1x + b0,
bn = bn−1 =··· = bm+1= 0. 數 ,
數 . f (x) + g(x)
f (x) + g(x) =∑ni=0(ai+ bi)xi. r∈ R, 數 r f (x) r f (x) =∑ni=0(rai)xi.
, 數 數 ,
數 P(R) . vectors
space 8 .
(1) f (x) =∑ni=0aixi, g(x) =∑ni=0bixi∈ P(R) f (x) + g(x) =
∑
n i=0(ai+ bi)xi=
∑
n i=0(bi+ ai)xi= g(x) + f (x).
(2) f (x) =∑ni=0aixi, g(x) =∑ni=0bixi, h(x) =∑ni=0cixi∈ P(R) ( f (x) + g(x)) + h(x) =
∑
n i=0(ai+ bi)xi+
∑
n i=0cixi=
∑
n i=0((ai+ bi) + ci)xi,
f (x) + (g(x) + h(x)) =
∑
n i=0aixi+
∑
n i=0(bi+ ci)xi=
∑
n i=0(ai+ (bi+ ci))xi.
(ai+ bi) + ci= ai+ (bi+ ci), ( f (x) + g(x)) + h(x) = f (x) + (g(x) + h(x)).
(3) g(x) = 0 =∑ni=0bixi∈ P(R), bi= 0, ∀i = 0,1,...,n.
f (x) =∑ni=0aixi∈ P(R), f (x) + g(x) =
∑
n i=0(ai+ bi)xi=
∑
n i=0aixi= f (x).
(4) f (x) =∑ni=0aixi∈ P(R), h(x) =∑ni=0(−ai)xi∈ P(R), f (x) + h(x) =
∑
n i=0(ai− ai)xi=
∑
n i=00xi= 0.
(5) r, s∈ R f (x) =∑ni=0aixi∈ P(R), r(s f (x)) = r(
∑
n i=0(sai)xi) =
∑
n i=0(r(sai))xi=
∑
n i=0((rs)ai)xi= (rs)
∑
n i=0aixi= (rs) f (x).
(6) r, s∈ R f (x) =∑ni=0aixi∈ P(R), (r + s) f (x) =
∑
n i=0((r + s)ai)xi=
∑
n i=0(rai+ sai)xi=
∑
n i=0(rai)xi+
∑
n i=0(sai)xi= r f (x) + s f (x).
(7) r∈ R f (x) =∑ni=0aixi, g(x) =∑ni=0bixi∈ P(R) r( f (x) + g(x)) = r(
∑
n i=0(ai+ bi)xi) =
∑
n i=0(r(ai+ bi))xi=
∑
n i=0(rai+ rbi)xi= r f (x) + rg(x).
(8) f (x) =∑ni=0aixi∈ P(R), 1 f (x) =
∑
n i=0(1ai)xi=
∑
n i=0aixi= f (x).
P(R) 數 vector space 8 , 數
P(R) vector space.
(C) L (Rn,Rm) Rn Rm linear transformations .
L (Rn,Rm) 數 , 數 L (Rn,Rm) (
Proposition 5.1.6). linear transformation 數
代 , Rm vector space, L (Rn,Rm)
數 vector space 8 . (2), T1, T2, T3∈ L (Rn,Rm).
v∈ Rn,
((T1+ T2) + T3)(v) = (T1+ T2)(v) + T3(v) = (T1(v) + T2(v)) + T3(v),
(T1+ (T2+ T3))(v) = T1(v) + (T2+ T3)(v) = T1(v) + (T2(v) + T3(v)).
Rm , ((T1+ T2) + T3)(v) = (T1+ (T2+ T3))(v).
v∈ Rn , 數 (T1+ T2) + T3= T1+ (T2+ T3).
(3) zero element zero mapping, T :Rn→ Rm, T (v) = O,
∀v ∈ Rn linear transformation. L (Rn,Rm) 數 vector space
8 , 數 L (Rn,Rm) vector space.
Rn , Rn Rn 數 vector
space, Rn subspace. vector space ,
vector space 數 vector space, vector space
subspace.
Definition 6.1.5. V vector space, W V nonempty subset. V
數 W vector space, W V subspace.
vector space subspace vector space, subspace
vector space 8 . vector space 8
(3)(4) , ,
. Rn subspace 性 ( Proposition 4.1.2).
Proposition 6.1.6. V vector space W V . W V subspace O∈ W u, v∈ W, r ∈ R u + rv∈ W.
Proof. (⇒) : subspace , 數 性, u, v∈ W, r ∈ R
u + rv∈ W. W , w∈ W. 0w, 性 0w∈ W.
V vector space, 0w = O (Proposition 6.1.3(2)). O =0w∈ W.
(⇐) : O∈ W, W V . u, v∈ W, r ∈ R
u + rv∈ W, V 數 W 性, W
數 . W V subspace, 數 W
vector space 8 . (3)(4) , V
. W V , W V , W
. O∈ W, (3) . w∈ W, 數 性 (−1)w ∈ W,
w + (−1)w = 0w = O, (4) .
Proposition 6.1.6, vector space V W V
subspace, (1) O∈ V
(2) u, v∈ V, r ∈ R ⇒ u + rv ∈ V.
. .
Example 6.1.7. (A) Mn×n, n× n vector space.
Mn×n symmetric matrices ( ) Mn×n subspace.
A∈ Mn×n symmetric matrix, AT= A. n× n O
symmetric matrix. A, B∈ Mn×n AT = A, BT = B, r ∈ R,
(A + rB)T = AT+ (rB)T= A + rB ( Proposition 3.2.4). A + rB symmetric
matrix, Mn×n symmetric matrices Mn×n subspace.
Mn×n invertible matrices ( ) Mn×n subspace
? . O invertible, O
Mn×n invertible matrices Mn×n subspace.
invertible matrices {O} , Mn×n subspace.
O , invertible matrices invertible. 2× 2
,
[ 1 0 0 1
] [
0 1 1 0
]
invertible,
[ 1 0 0 1
] +
[ 0 1 1 0
]
=
[ 1 1 1 1
]
invertible.
(B) P(R), x 數 數 vector space.
數 n∈ N, 數 n Pn(R) P(R) subspace.
Pn(R) Pn(R) = {∑ni=0aixi| ai∈ R}. Pn(R)
( 數學 數 −∞, 0. 代數
, ). f (x) =∑ni=0aixi, g(x) =∑ni=0bixi∈ Pn(R), r∈ R, f (x) + rg(x) =∑ni=0(ai+ rbi)xi∈ Pn(R). Pn(R) P(R) subspace. ,
數 n , P(R) subspace .
. , 數 n 數
, (x2+ x + 1) + (−x2+ x + 1) = 2x + 2. ,
vector space.
(C) L (Rn,Rm) Rn Rm linear transformations vector space.
Rn subspace V ,
N(V ) ={T ∈ L (Rn,Rm)| V ⊆ ker(T)}.
N(V ) v∈ V, T(v) = O linear transformation T .
, L (Rn,Rm) additive identity, zero mapping, kernel Rn V ⊆ Rn
zero mapping N(V ). T1, T2∈ N(V), r∈ R, v∈ V
(T1+ rT2)(v) = T1(v) + r(T2(v)) = O + rO = O.
V ⊆ ker(T1+ rT2), T1+ rT2∈ N(V). N(V ) L (Rn,Rm) subspace.
Question 6.2. L (Rn,Rm) Rn Rm linear transformations vector
space. Rn subspace V ,
S(V ) ={T ∈ L (Rn,Rm)| ker(T) ⊆ V}.
S(V ) L (Rn,Rm) subspace?
Rm subspace W ,
R(W ) ={T ∈ L (Rn,Rm)| T(Rn)⊆ W}.
R(W ) L (Rn,Rm) subspace?
I(W ) ={T ∈ L (Rn,Rm)| W ⊆ T(Rn)}.
I(W ) L (Rn,Rm) subspace?
6.2. Basis and Dimension
, vector space basis , finitely generated vector
space dimension 性 .
Rn subspace , basis . basis, subspace
, basis 線性 . basis subspace
spanning vectors linearly independent ( Proposition 4.1.8).
vector space.
Definition 6.2.1. V vector space v1, . . . , vn∈ V. c1, . . . , cn∈ R, c1v1+···+cnvn v1, . . . , vn linear combination. v1, . . . , vn linear combination
, Span(v1, . . . , vn) , Span(v1, . . . , vn) ={∑ni=1civi| c1, . . . , cn∈ R}.
Span(v1, . . . , vn) V subspace ( Proposition 1.3.2).
v1, . . . , vn subspace. W V subspace v1, . . . , vn∈ W, subspace 數 性 Span(v1, . . . , vn)⊆ W. , Span(v1, . . . , vn) = V
, .
Definition 6.2.2. V vector space. v1, . . . , vn∈ V Span(v1, . . . , vn) = V , {v1, . . . , vn} V spanning set. V finitely generated vector space.
finitely generated vector space ? vector space
線性 . Rn finitely generated ( e1, . . . en
). 初 Rn subspaces, finitely generated vector space
. vector space, finitely generated,
, .
Example 6.2.3. 前 vector space finitely generated vector space.
(A) Mm×n finitely generated. Mi, j∈ Mm×n, (i, j)-th entry 1,
entry 0 m× n matrix. m× n matrix Mi, j 1≤ i ≤ m,
1≤ j ≤ n linear combination. {Mi, j| 1 ≤ i ≤ m,1 ≤ j ≤ n} Mm×n spanning set, Mm×n finitely generated vector space.
(B) P(R) finitely generated vector space. { f1(x), . . . , fn(x)} P(R) spanning set, f1(x), . . . , fn(x) m, f1(x), . . . , fn(x) linear combi- nation c1f1(x) +··· + cnfn(x) 數 大 m. Span( f1(x), . . . , fn(x))
數大 m . { f1(x), . . . , fn(x)} P(R) spanning set ,
P(R) finitely generated. 數 n Pn(R)
finitely generated vector space. {xn, . . . , x, 1} Pn(R) spanning set.
大 finite generated vector space subspace finitely gener-
ated. , (大 ).
linearly independence , .
Spanning set linear combination 性, linear independence
linear combination 性. Rn linearly independent
vector space.
Definition 6.2.4. V vector space v1, . . . , vn∈ V. c1, . . . , cn∈ R 0 c1v1+···+cnvn= O, v1, . . . , vn linearly dependent. , c1, . . . , cn
0 c1v1+··· + cnvn= O, v1, . . . , vn linearly independent.
Rn , v1, . . . , vn linearly independent, c1, . . . , cn c1v1+··· + cnvn. v1, . . . , vn linear combination
. , v1, . . . , vn linearly dependent, d1, . . . , dn 0 d1v1+···+cnvn= O. c1, . . . , cn∈ R, c1v1+···+cnvn (c1+ d1)v1+···+(cn+ dn)vn
v1, . . . , vn linear combination, .
(c1+ d1)v1+··· + (cn+ dn)vn=
(c1v1+··· + cnvn) + (d1v1+··· + dnvn) = c1v1+··· + cnvn+ O = c1v1+··· + cnvn. Example 6.2.5. P(R) xn, xn−1, . . . , x, 1 linearly independent.
cn, . . . , c1, c0 ∈ R cnxn+··· + c1x + c01 , ,
cn, . . . , c1, c0 0. P(R) linearly independent
, Lagrange interpolation polynomials. , 大
.
a, b, c 數, p1(x), p2(x), p3(x)
p1(a) = 1, p1(b) = p1(c) = 0, p2(b) = 1, p2(a) = p2(c) = 0 and p3(c) = 1, p3(a) = p3(b) = 0.
p1(b) = p1(c) = 0, p1(x) (x− b)(x − c) , 數 r p1(x) = r(x− b)(x − c). p1(a) = 1, 代 x = a r = 1/(a− b)(a − c).
p2(x), p3(x)
p1(x) = (x− b)(x − c)
(a− b)(a − c), p2(x) = (x− a)(x − c)
(b− a)(b − c) and p3(x) = (x− a)(x − b) (c− a)(c − b).
p1(x), p2(x), p3(x) linearly independent. , f (x) = c1p1(x) + c2p2(x) + c3p3(x), 代 x = a p1(a) = 1, p2(a) = p3(a) = 0, f (a) = c1.
f (b) = c2, f (c) = c3. f (x) , f (a) = f (b) = f (c) = 0, c1= c2= c3= 0. c1= c2= c3= 0 c1p1(x) + c2p2(x) + c3p3(x)
, p1(x), p2(x), p3(x) linearly independent.
. n 數 a1, . . . , an, n
n− 1 p1(x), . . . , pn(x) pi(ai) = 1 j̸= i , pi(aj) = 0. 前 , p1(x), . . . , pn(x) linearly independent.
Spanning set linearly independent 性 ,
. .
Lemma 6.2.6. V vector space v1, . . . , vn∈ V.
(1) Span(v1, . . . , vn−1)̸= Span(v1, . . . , vn−1, vn) vn+1̸∈ Span(v1, . . . , vn).
(2) v1, . . . , vn−1 linearly independent, v1, . . . , vn−1, vn linearly indepen- dent vn̸∈ Span(v1, . . . , vn−1).
(3) w1, . . . , wm∈ V. w1, . . . , wm∈ Span(v1, . . . , vn) m > n, w1, . . . , wm linearly dependent.
Proof. (1), (2), (3) Lemma 4.2.1, Lemma 4.2.4 Lemma 4.2.5 vector space . (1), (2) Lemma 4.2.1, Lemma 4.2.4 .
(3) . Lemma 4.2.5 Rm ,
vector space, .
w1, . . . , wm∈ Span(v1, . . . , vn), j = 1, . . . , m, wj v1, . . . , vn
linear combination. , a1, j, . . . , ai, j, . . . , an, j∈ R wj= a1, jv1+··· + ai, jvi+··· + an, jvn.
c1, . . . , cm∈ R 0 c1w1+··· + cmwm= O, w1, . . . , wm linearly dependent. c1w1+···+cmwm wj v1, . . . , vn linear combination
(c1a1,1+··· + cma1,m)v1+··· + (c1ai,1+··· + cmai,m)vi+··· + (c1am,1+··· + cmam,m)vm. (6.1)
c1, . . . , cm ∈ R (6.1) vi 數 0,
c1w1+··· + cmwm= O.
a1,1x1+··· + a1,mxm = 0 ...
ai,1x1+··· + ai,mxm = 0 ...
an,1x1+··· + an,mxm = 0
x1= c1, . . . , xm= cm, c1w1+··· + cmwm= O. homogeneous
linear system 數 n 數 數 m, Corollary 3.4.7
0 c1, . . . , cm∈ R x1= c1, . . . , xm= cm . w1, . . . , wm linearly
dependent.
finitely generated vector space subspace finitely generated.
Proposition 6.2.7. V finitely generated vector space. W V subspace, W finitely generated vector space.
Proof. V finitely generated , v1, . . . , vn∈ V Span(v1, . . . , vn) = V .
{O} = Span(O) finitely generated, W̸= {O} . ,
W finitely generated. w1∈ W w1̸= O. W finitely generated, Span(w1)̸= W, w2∈ W w2̸∈ Span(w1). Lemma 6.2.6 (2) w1, w2 linearly independent. , W finitely generated, Span(w1, w2)̸= W
w3∈ W w3̸∈ Span(w1, w2). Lemma 6.2.6 (2) w1, w2, w3 linearly independent.
, 數學 w1, . . . , wk∈ W linearly independent.
Span(w1, . . . , wk)̸= W, wk+1∈ W wk+1̸∈ Span(w1, . . . , wk). Lemma 6.2.6 (2) w1, . . . , wk, wk+ linearly independent. 數學 , W finitely generated, m∈ N, w1, . . . , wm∈ W linearly independent. m > n
. w1, . . . , wm∈ W ⊆ V = Span(v1, . . . , vn), Lemma 6.2.6 (3) w1, . . . , wm linearly dependent. W finitely generated vector space. Question 6.3. P(R) n∈ N, xn, xn−1, . . . , x, 1 linearly independent,
P(R) finitely generated vector space.
spanning set 性 linearly independent 性, {v1, . . . , vn} vector space V spanning set linearly independent, V
v1, . . . , vn linear combination, .
Definition 6.2.8. V vector space. {v1, . . . , vn} V spanning set linearly independent, v1, . . . , vn V basis.
finitely generated vector space basis. 性 finitely generated vector space ,
vector space finitely generated, .
Theorem 6.2.9. V ̸= {O} finitely generated vector space. v1, . . . , vn∈ V V basis. w1, . . . , wm∈ V V basis, m = n.
Proof. Theorem 4.3.1 Theorem 4.3.2 vector space .
Theorem 4.3.1 basis 性. finitely generated
性 , 數學 .
vector space spanning set 數 數學 . V
. V = Span(u). V ̸= {O}, u̸= O. u linearly
independent {u} V spanning set u V basis. vector
space k , basis . V k + 1 vector
space, V = Span(u1, . . . , uk, uk+1). W = Span(u1, . . . , uk). W
k vector space, v1, . . . , vn∈ W W basis.
Span(v1, . . . , vn) = W v1, . . . , vn linearly independent. W = V , v1, . . . , vn
V basis. W̸= V, uk+1̸∈ W ( V = Span(u1, . . . , uk, uk+1)⊆ W
). uk+1̸∈ W = Span(v1, . . . , vn) Lemma 6.2.6 (2) v1, . . . , vn, uk+1 linearly independent. V = Span(u1, . . . , uk, uk+1) Span(u1, . . . , uk) = W = Span(v1, . . . , vn)
V = Span(v1, . . . , vn, uk+1). v1, . . . , vn, uk+1 V basis.
basis 數 , Lemma 4.2.5 Theorem 4.3.2
vector space , Rm . ,
.
Theorem 6.2.9 V basis 數 . n
V basis, V basis n .
.
Definition 6.2.10. V finitely generated vector space. V basis
數 V dimension ( ), dim(V ) .
finitely generated vector space basis 數 ,
finitely generated vector space finite dimensional vector space.
Example 6.2.11. Example 6.2.3 finite dimensional vector space .
(A) Example 6.2.3 (A) {Mi, j∈ Mm×n| 1 ≤ i ≤ m,1 ≤ j ≤ n} Mm×n
spanning set. linearly independent, {Mi, j| 1 ≤ i ≤ m,1 ≤ j ≤ n}
Mm×n basis. dim(Mm×n) = m× n.
(B) {xn, . . . , x, 1} Pn(R) spanning set linearly independent.
xn, . . . , x, 1 Pn(R) basis, dim(Pn(R)) = n + 1.
finite dimensional vector space dimension 性 , .
Proposition 4.3.4, Corollary 4.3.6.
Proposition 6.2.12. V finite dimensional vector space.
(1) {v1, . . . , vn} V spanning set, dim(V )≤ n. , v1, . . . , vn linearly dependent, dim(V ) < n.
(2) v1, . . . , vn∈V linearly independent, dim(V )≥ n. , {v1, . . . , vn} V spanning set, dim(V ) > n.
(3) W V subspace, dim(W )≤ dim(V). dim(W ) = dim(V ) V = W .
(4) v1, . . . , vn∈ V. .
(a) v1, . . . , vn V basis.
(b) dim(V ) = n {v1, . . . , vn} V spanning set.
(c) dim(V ) = n v1, . . . , vn linearly independent.
, Proposition 6.2.12 (4) v1, . . . , vn V basis ,
dim(V ) n, spanning set linearly independent
.
Example 6.2.13. Example 6.2.5 a, b, c 數, p1(x) = (x− b)(x − c)
(a− b)(a − c), p2(x) = (x− a)(x − c)
(b− a)(b − c) and p3(x) = (x− a)(x − b) (c− a)(c − b).
p1(a) = 1, p1(b) = p1(c) = 0; p2(b) = 1, p2(a) = p2(c) = 0; p3(c) = 1, p3(a) = p3(b) = 0.
p1(x), p2(x), p3(x)∈ P2(R) linearly independent dim(P2(R)) = 3, Propo- sition 6.2.12 (4) p1(x), p2(x), p3(x) P2(R) basis.
n 數 a1, . . . , an, n n−1 p1(x), . . . , pn(x) pi(ai) = 1 j̸= i , pi(aj) = 0. p1(x), . . . , pn(x)∈ Pn−1(R) linearly independent,
dim(Pn−1(R)) = n p1(x), . . . , pn(x) Pn−1(R) basis.
6.3. Linear Transformation
vector space linear transformation 性 . Rn
, vector spaces , 數 vector space
數 , .
Definition 6.3.1. V,W vector spaces, T : V → W 數. T v1, . . . , vk∈ V c1, . . . , ck∈ R
T (c1v1+··· + ckvk) = c1T (v1) +··· + ckT (vk).
T linear transformation. T linear.
c1v1+··· + cnv1 V linear combination, c1T (v1) +··· + ckT (vk)
W linear combination, . O∈ V V zero element ,
linear transformation , T (O) = T (O + O) = T (O) + T (O). T (O)
, T (O) W zero element. linear transformation
T : V→ W, V zero element W zero element. V̸= W ,
zero element , O , .
T (O) = O .
T : V→ W linear transformation, V linear
combination 代 T linear transformation . Rn (
Proposition 5.1.3), linear combination .
Proposition 6.3.2. V,W vector spaces T : V → W 數. T linear transformation u, v∈ V, r ∈ R T (u + rv) = T (u) + rT (v).
T1, T2 V W linear transformation , T1, T2
T1+ T2: V → W, (T1+ T2)(v) = T1(v) + T2(v), ∀v ∈ V. linear transformation T : V → W 數 . r∈ R, rT : V → W (rT )(v) = r(T (v)), ∀v ∈ V.
T1+ T2 rT V W linear transformation ( Proposition 5.1.6).
數 vector space 8 , (3) additive
identity V W zero mapping ( V W O).
( ).
Proposition 6.3.3. V,W vector spaces, L (V,W) V W linear
transformation . 數 數 , L (V,W) vector
space.
linear transformation “ 數”. U,V,W vector
spaces, T : V → W T′: W → U linear transformations, T′◦ T : V → U T′◦ T(v) = T′(T (v)), ∀v ∈ V. ( Proposition 5.1.7).
Proposition 6.3.4. U,V,W vector spaces. T : V → W T′: W → U linear transformations, T′◦ T : V → U linear transformation.
Rn Rm linear transformation standard matrix representation
linear transformation basis . linear
transformation ( Theorem 5.1.8).
Theorem 6.3.5. V,W vector spaces v1, . . . , vn∈ V, V basis.
w1, . . . , wn∈ W, linear transformation T : V → W, i = 1, . . . , n T (vi) = wi.
linear transformation 數 性, subspace .
T : V → W linear transformation , V subspace V′ T (V′) ={T(v) ∈ W | v ∈ V′} = {w ∈ W | w = T(v), for some v ∈ V′}.
W′ W subspace,
T−1(W′) ={v ∈ V | T(v) ∈ W′}.
( Proposition 5.2.1).
Proposition 6.3.6. V,W vector spaces T : V → W linear transformation.
V′ V subspaces, T (V′) W subspace. , W′ W subspaces,
T−1(W′) V subspace.
, V′= V W′={O} ,
T (V ) ={w ∈ W | w = T(v) for some v ∈ V} and T−1({O}) = {v ∈ V | T(v) = O}
subspaces, T linear transformation . Rn .
Definition 6.3.7. V,W vector spaces T : V → W linear transformation.
W subspace T (V ) T range ( image). V subspace T−1({O})
T kernel, ker(T ) .
, linear transformation T : V → W onto range T (V ) W .
T range ? ( Proposition 5.2.4).
Proposition 6.3.8. V,W vector spaces T : V → W linear transformation.
{v1, . . . , vn} V spanning set.
T (V ) = Span(T (v1), . . . , T (vn)).
, T onto W = Span(T (v1), . . . , T (vn)).
kernel one-to-one . Proposition 5.2.6
.
Proposition 6.3.9. V,W vector spaces T : V → W linear transformation.
T one-to-one ker(T ) ={O}.
Proposition 6.3.8 T : V→ W onto spanning set. T
V spanning set W spanning set. linear transformation
linearly independent ? .
Proposition 6.3.10. V,W vector spaces T : V → W one-to-one linear transformation. v1, . . . , vn∈ V linearly independent T (v1), . . . , T (vn)∈ W linearly independent.
Proof. (⇒) v1, . . . , vn∈ V linearly independent, T (v1), . . . , T (vn)∈ W linearly independent. , T (v1), . . . , T (vn)∈ W linearly dependent,
c1, . . . , cn∈ R 0 c1T (v1) +··· + cnT (vn) = O. T linear, O = c1T (v1) +··· + cnT (vn) = T (c1v1+··· + cnvn). T one-to-one, c1v1+··· + cnvn= O T (c1v1+··· + cnvn) = O. v1, . . . , vn∈ V linearly independent c1=··· = cn= 0. c1, . . . , cn∈ R 0 ,
T (v1), . . . , T (vn)∈ W linearly independent.
(⇐) T one-to-one, T linear transformation.
T (v1), . . . , T (vn)∈ W linearly independent. v1, . . . , vn∈ V linearly dependent, c1, . . . , cn∈ R 0 c1v1+···+cnvn= O. T linear,
c1T (v1) +··· + cnT (vn) = T (c1v1+··· + cnvn) = T (O) = O. T (v1), . . . , T (vn)∈ W linearly independent, c1=··· = cn= 0 . v1, . . . , vn∈ V linearly
independent.
T : V → W one-to-one onto , invertible, T−1: W→ V, v∈ V T−1◦ T(v) = T−1(T (v)) = v w∈ W T◦ T−1(w) =
T (T−1(w)) = w. T−1 T inverse. Theorem 5.3.12 Rn
linear transformation invertible, inverse linear transformation.
, Theorem 5.3.12 standard matrix representation
. matrix representation, .
Theorem 6.3.11. V,W vector spaces T : V→ W linear transformation.
T invertible, T inverse T−1: W → V linear transformation.
Proof. w∈ W, T one-to-one onto, v∈ V T (v) = w.
數 T−1(w) = v. T−1 W V 數.
T−1: W → V linear transformation. w1, w2∈ V, r ∈ R,
T−1(w1+ rw2) = T−1(w1) + rT−1(w2). T−1(w1) = v1, T−1(w2) = v2. T (v1) = w1
T (v2) = w2. T−1(w1+ rw2) T−1(w1) + rT−1(w2) = v1+ rv2
T (v1+ rv2) = w1+ rw2. T linear transformation, T (v1+ rv2) = T (v1) + rT (v2) = w1+ rw2. T−1(w1+ rw2) = v1+ rv2= T−1(w1) + rT−1(w2),
T−1 linear transformation.
T : V→ W invertible linear transformation , T isomor-
phism. V W vector space T
數. T onto, T V W spanning set, T one-to-one
T linearly independent , .
Theorem 6.3.12. V,W vector spaces T : V → W isomorphism. v1, . . . , vn V basis T (v1), . . . , T (vn) W basis.
Proof. (⇒) v1, . . . , vn V basis. T onto {v1, . . . , vn} V spanning set, Proposition 6.3.8 {T(v1), . . . , T (vn)} W spanning set.
T one-to-one v1, . . . , vn ∈ V linearly independent, Proposition 6.3.10 T (v1), . . . , T (vn)∈ W linearly independent. T (v1), . . . , T (vn) W basis.
(⇐) T (v1), . . . , T (vn) W basis. T−1: W→ V linear transformation (Theorem 6.3.11), T : V → W inverse, T−1 one-to-one onto,
T−1: W → V isomorphism. T (v1), . . . , T (vn) W basis, 前
T−1(T (v1)), . . . , T−1(T (vn)) V basis. i = 1, . . . , n T−1(T (vi)) = vi,
v1, . . . , vn V basis.
V,W finite dimensional vector space , Theorem 6.3.12, T : V → W
isomorphism, V W basis 數 , dim(V ) = dim(W ).
dim(V )̸= dim(W), V W isomorphism. 性
, .
Corollary 6.3.13. V,W finite dimensional vector spaces. T : V → W isomorphism dim(V ) = dim(W ).
Proof. (⇒) v1, . . . , vn V basis, dim(V ) = n. T : V → W isomorphism Theorem 6.3.12 T (v1), . . . , T (vn) W basis. dim(W ) = n = dim(V ).
(⇐) dim(V ) = dim(W ), v1, . . . , vn w1, . . . , wn V basis W basis. linear transformation T : V→W T (vi) = wi,∀i = 1,...,n ( Theo- rem 6.3.5). T isomorphism. 6.3.8 T (V ) = Span(T (v1), . . . , T (vn)) = Span(w1, . . . , wn). w1, . . . , wn W basis Span(w1, . . . , wn) = W .
T (V ) = W , T onto. v∈ ker(T). v1, . . . , vn V basis, c1, . . . , cn∈ R v = c1v1+··· + cnvn. T (v) = O,
O = T (c1v1+··· + cnvn) = c1T (v1) +··· + cnT (vn) = c1w1+··· + cnwn.
w1, . . . , wn linearly independent, c1=··· = cn= 0. v = c1v1+···+cnvn= O.
ker(T ) ={O}, Proposition 6.3.9 T one-to-one, T isomorphism.
6.4. Coordinatization
, linear transformation, vector space
. vector space , Rn
.
V finite dimensional vector space, V basis, basis
, , basis, ordered
basis. , basis , ordered
basis.([ ordered basis , (v1, . . . , vn) , .
1 0 ]
, [0
1
]) ([
0 1 ]
, [1
0 ])
R2 ordered basis.
, ordered basis , ordered
basis. B = (v1, . . . , vn) V ordered basis, B
ordered basis (v1, . . . , vn). Rn standard basis, E (e1, . . . , en) ordered basis.
vector space V ordered basis B = (v1, . . . , vn) , V
“ ” (coordinatization). v∈ V, B ordered basis
v v = c1v1+··· + cnvn ,
c1
... cn
B v .
, B[v] B v . ,
B[v] Rn . vector space ,
Rn .
Example 6.4.1. 前 vector space .
(A) M2×2 ordered basis E =
([ 1 0 0 0
] ,
[ 0 1 0 0
] ,
[ 0 0 1 0
] ,
[ 0 0 0 1
])
( basis M2×2 standard basis). M2×2
[ a b c d
] , [ a b
c d ]
= a [ 1 0
0 0 ]
+ b [ 0 1
0 0 ]
+ c [ 0 0
1 0 ]
+ d [ 0 0
0 1 ]
, [ a b
c d ]
E
E
[ a b c d
]
=
a b c d
.
E
[ 1 −2
−3 4
]
=
1
−2−3 4
.
(B) P2(R) x2, x, 1 basis standard basis. E = (x2, x, 1) ordered basis. 2x2− 3x + 4 E
2
−3 4
,
E[2x2− 3x + 4] =
2
−3 4
.
ordered basis B = (p1(x), p2(x), p3(x))
p1(x) =−(x − 1)(x + 1), p2(x) = (1/2)x(x + 1) and p3(x) = (1/2)x(x− 1) ( Example 6.2.5, Example 6.2.13).
p1(0) = 1, p1(1) = p1(−1) = 0; p2(1) = 1, p2(0) = p2(−1) = 0; p3(−1) = 1, p3(0) = p3(1) = 0, 2x2−3x+4 = c1p1(x) + c2p2(x) + c3p3(x), 代 x = 0, 1,−1, c1= 4, c2= 3, c3= 9.
B[2x2− 3x + 4] =
4 3 9
.