-
Notifications
You must be signed in to change notification settings - Fork 0
/
mla.tex
263 lines (231 loc) · 8.95 KB
/
mla.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
\section{Lineare und Multilineare Algebra}
\label{sec:mla}
\textbf{Vereinbarung:} In diesem Abschnitt sei $\K=\R,\C$ (also ein
vernünftiger Körper) und alle
auftretenden Vektorräume (bis auf die frei erzeugten) seien
endlichdimensionale $\K$-Vektorräume. Insbesondere sollen
$V_1,\dots,V_n, U,V,W$ endlichdimensionale $\K$-Vektorräume bezeichnen
\subsection{Vorbemerkungen (aka Erinnerungen an die Grundvorlesungen)}
\label{sec:vbakaegv}
Wir geben hier eine Sammlung von Begriffen und ihren Eigenschaften,
die man kennen sollte.
\begin{definition}
Sei $M$ eine nichtleere Menge. Versieht man die Menge
\begin{equation*}
F(M) := \set{f \colon M \to \K \mid \supp f\text{ endlich}}
\end{equation*}
mit der kanonischen Vektorraumstruktur für $\K$-wertige Abbildungen,
so heißt $F(M)$ der \emph{frei von $M$ erzeugte Vektorraum}.
Die kanonische Einbettung von $M$ in $F(M)$ bezeichnen wir mit
\begin{equation*}
\delta \colon M \to F(M),\, x \mapsto \delta_x
\end{equation*}
wobei $\delta_x(y) := 1$ falls $x=y$ und $0$ sonst.
\end{definition}
\begin{proposition}
Es gilt
\begin{statements}
\item Die Menge $\delta(M)$ bildet eine Basis von $F(M)$.
\item \textit{(Universelle Eigenschaft von $(F(M),\delta)$)} Ist
$f\colon M \to U$ eine Abbildung, so gibt es genau eine lineare
Abbildung $F\colon F(M) \to U$, sodass $f = F\circ\delta$.
\end{statements}
\end{proposition}
\begin{proposition}[Universelle Eigenschaft des Quotienten]
Sei $U$ ein Unterraum von $V$. Ist $f\colon
V \to W$ eine lineare Abbildung, sodass $U \subset \ker f$, dann
gibt es eine eindeutig bestimmte lineare Abbildung $F\colon V/U \to
W$, sodass $f = F\circ \pr$, wobei $\pr\colon V \to V/U$ die
kanonische Projektion bezeichnet.
\end{proposition}
\begin{definition}
Sei $b$ eine bilineare Abbildung $b\colon V\times W \to U$.
Man nennt $b$ \emph{nichtentartet}, falls
\begin{equation*}
\forall v\in V\colon b(v,w) = 0 \implies w=0\text{ und }
\forall w\in W\colon b(v,w) = 0 \implies v=0
\end{equation*}
Man nennt $b$ eine \emph{Paarung}, falls $U=\K$ und eine
nichtentartete Paarung heißt auch \emph{Dualität}.
\end{definition}
\begin{proposition}
Eine Dualität liefert je einen Isomorphismus $\Phi\colon V\to W^*$
und $\Psi\colon W\to V^*$.
\end{proposition}
\subsection{Tensorprodukt}
\label{sec:tensor}
\begin{definition}
Seien $V,W$ Vektorräume und bezeichne $I(V\times W)$ den von den Elementen
\begin{equation*}
\delta((\lambda v, w)) - \lambda \delta((v,w)),\,
\delta((v, \lambda w)) - \lambda \delta((v,w)),\,
\delta((v + v', w)) - \delta((v,w)) -\delta((v',w)),\,
\delta((v, w + w')) - \delta((v,w)) -\delta((v,w'))
\end{equation*}
erzeugten Vektorunterraum von $F(V\times W)$, wobei $v,v'\in V$,
$w,w'\in W$ und $\lambda\in\K$. Dann ist das \emph{Tensorprodukt von
$V$ und $W$ über $\K$} definiert als
\begin{equation*}
V\tensor_\K W := V \tensor W := F(V\times W)/I(V\times W).
\end{equation*}
Wir bezeichnen mit $\tau\colon V\times W \to V\tensor W, \tau((v,w))
:= v\tensor w := \pr\circ\delta((v,w))$ die kanonische Einbettung
von $V\times W$ in $V\tensor W$, wobei $\pr\colon F(V\times W) \to
F(V\times W)/I(V\times W)$ die kanonische Projektion ist.
\end{definition}
\begin{proposition}
Seien $U,V,W$ Vektorräume. Dann gilt
\begin{statements}
\item Die Abbildung $\tau$ ist bilinear.
\item Die Menge $\tau(V\times W)$ bildet ein Erzeugendensystem von
$V\tensor W$.
\item \textit{(Universelle Eigenschaft des Tensorprodukts)} Sei $b\in
\Mult(V,W;U)$. Dann gibt es eine eindeutig bestimmte lineare
Abbildung $B\colon V\tensor W \to U$, sodass $B\circ\tau = b$.
\end{statements}
\end{proposition}
\begin{proposition}
Seien $U,V,W$ Vektorräume und $b\in\Mult(V,W;U)$. Dann gilt
\begin{statements}
\item $V^*\tensor W \equiv \Hom(V,W)$. Insbesondere gilt $\dim
V\tensor W = \dim V \cdot \dim W$.
\item $V\tensor W \equiv W\tensor V$
\item $V\tensor\K \equiv V$
\item\label{tp:asso} $U\tensor(V\tensor W) \equiv (U\tensor V)\tensor W$
\item $U\tensor(V\oplus W) \equiv U\tensor V \oplus U\tensor W$
\end{statements}
\end{proposition}
\begin{bemerkung}
Aus \ref{tp:asso} folgt die Wohldefiniertheit (bis auf Isomorphie)
eines mehrfachen Tensorprodukts. Wir können festlegen, dass
\begin{equation*}
V_1\tensordots V_n := V_1\tensor (V_2\tensordots V_n).
\end{equation*}
Wir erhalten eine multilineare Abbildung
\begin{equation*}
\tilde\tau_n \colon V_1\dotted{\times}V_n \to
V_1\tensordots V_n, (v^1,\dots,v^n) \mapsto \tau(v^1,\tilde\tau_{n-1}(v^2,\dots, v^n))
\end{equation*}
die wir im folgenden auch einfach mit $\tau$ bezeichnen. Es gilt
wie vorher auch, dass die Elemente $v^1\tensordots v^n :=
\tau(v^1,\dots,v^n)$ ein Erzeugendensystem bilden. Die
universelle Eigenschaft erweitert sich wie folgt:
\end{bemerkung}
\begin{proposition}[Universelle Eigenschaft des mehrfachen Tensorprodukts]
Seien $V_1,\dots,V_n,W$ Vektorräume und $m\in
\Mult(V_1,\dots,V_n;W)$. Dann gibt es eine eindeutig bestimmte
lineare Abbildung $M\colon V_1\tensordots V_n \to W$, sodass $m =
M\circ \tau$.
\end{proposition}
\begin{proposition}
Seien $V_1,\dots,V_n,W$ Vektorräume. Dann gilt
\begin{equation*}
(V_1\tensordots V_n)^* \equiv V_1^*\tensordots V_n^*
\end{equation*}
und
\begin{equation*}
V_1^*\tensordots V_n^*\tensor W \equiv \Mult(V_1,\dots,V_n;W)
\end{equation*}
\end{proposition}
\subsection{Tensoralgebra und äußere Algebra}
\label{sec:tauv}
\begin{definition}
Wir definieren
\begin{equation*}
T^0(V) := \K,\, T^n(V) := V\tensor T^{n-1}(V)\text{ und }
T^{(p,q)}(V) := T^p(V) \tensor T^q(V^*).
\end{equation*}
und identifizieren kanonisch $T^{(p,0)}(V)$ mit $T^p(V)$ und
$T^{(0,p)}(V)$ mit $T^p(V^*)$.
Es heißt
\begin{equation*}
T(V) := \bigoplus_{k\geq 0} T^k(V)
\end{equation*}
die \emph{Tensoralgebra von $V$} und
\begin{equation*}
\mathcal{T}(V) := \bigoplus_{p,q \geq 0} T^{(p,q)}(V)
\end{equation*}
die \emph{erweiterte Tensoralgebra von $V$}.
Wir definieren $\tensor\colon \mathcal{T}(V)\times \mathcal{T}(V)
\to \mathcal{T}(V)$ als die bilineare Abbildung, die eindeutig durch
\begin{equation*}
(v^1\tensordots v^k\tensor \nu_1\tensordots \nu_l,
w^1\tensordots w^p\tensor \mu_1\tensordots \mu_q) \mapsto
v^1\tensordots v^k\tensor w^1\tensordots w^p \tensor
\nu_1\tensordots \nu_l \tensor \mu_1 \tensordots \mu_q
\end{equation*}
bestimmt ist.
\end{definition}
\begin{proposition}
Die erweiterte Tensoralgebra ist eine $\Z$-graduierte kommutative $\K$-Algebra
mit $1$. Die Tensoralgebra ist eine $\Z$-graduierte Unteralgebra der
erweiterten Tensoralgebra.
\end{proposition}
\begin{definition}
Bezeichne $S_k$ die Menge der Permutationen der Menge
$\set{1,\dots,k}$. Eine Permutation $\sigma \in S_k$ induziert eine
lineare Abbildung $\sigma \colon T^k(V) \to T^k(V)$, die durch
\begin{equation*}
v^1\tensordots v^k \mapsto v^{\sigma(1)}\tensordots v^{\sigma(k)}
\end{equation*}
eindeutig definiert ist.
Wir definieren
\begin{equation*}
\Lambda^k(V) := \set{ t\in T^k(V) \mid \sigma(t) =
\sgn(\sigma)\cdot t }
\end{equation*}
die Menge der \emph{alternierenden Tensoren vom Grad $k$} und die
\emph{äußere Algebra von $V$} ist definiert als
\begin{equation*}
\Lambda(V) := \bigoplus_{k\geq 0}\Lambda^k(V)
\end{equation*}
Weiters bezeichne
\begin{equation*}
\Alt_k \colon T^k(V) \to T^k(V), t \mapsto
\begin{cases}
\frac{1}{k!}\sum_{\sigma\in S_k} \sgn(\sigma) \sigma(t) & 0\leq
k \leq \dim V \\
0 & \sonst
\end{cases}
\end{equation*}
den (manchmal) sogenannten \emph{Alterator}.
\end{definition}
\begin{proposition}
Sei $\sigma \in S_k$ eine Permutation. Es gilt
\begin{statements}
\item $\sigma(\Alt_k(t)) = \Alt_k(\sigma(t)) =
\sgn(\sigma)\Alt_k(t)$.
\item $\Alt_k \colon T^k(V) \to \Lambda^k(V)$ ist eine surjektive
Projektion. Insbesondere sind also Elemente der Form
\begin{equation*}
v^1\dotted{\wedge} v^k := k!\Alt_k(v^1\tensordots v^k)
\end{equation*}
für $v^1,\dots,v^k\in V$ ein Erzeugendensystem für $\Lambda^k(V)$.
\item $\dim \Lambda^k(V) = \binom{\dim V}{k}$
\end{statements}
\end{proposition}
\begin{definition}
Wir definieren das \emph{äußere Produkt} als die bilineare Abbildung
$\Lambda(V\times \Lambda(V) \to \Lambda(V)$, die durch
\begin{equation*}
(v^1\wedgedots v^k) \wedge (w^1\wedgedots w^l) :=
\frac{(k+l)!}{k!l!}\Alt_{k+l}((v^1\wedgedots v^k) \tensor
(w^1\wedgedots w^l))
\end{equation*}
eindeutig bestimmt ist.
\end{definition}
\begin{proposition}
Es gilt
\begin{statements}
\item Das äußere Produkt ist assoziativ.
\item Das äußere Produkt ist ``superkommuativ'', das heißt für
$\omega \in \Lambda^k(V)$ und $\eta \in \Lambda^l(V)$ gilt
\begin{equation*}
\omega \wedge \eta = (-1)^{kl}\eta\wedge \omega
\end{equation*}
\end{statements}
\end{proposition}
%%% Local Variables:
%%% mode: latex
%%% TeX-master: "main"
%%% End: