Skip to content

Commit aea731f

Browse files
authored
update for WS23/24 (#9)
* update for WS23/24 Additions: spezielle orthogonale Matrizen, Kompression, Rekursive Folgen, Definitheit & Quadratische Funktionen Changes: minor additions and changes in 1., 2.3, 4.5, 8., 9., 10., 11.2 * add more understandable description in 2.8
1 parent 687c757 commit aea731f

File tree

1 file changed

+114
-34
lines changed

1 file changed

+114
-34
lines changed

Lineare-Algebra.tex

Lines changed: 114 additions & 34 deletions
Original file line numberDiff line numberDiff line change
@@ -98,8 +98,7 @@ \subsection{Allgemeines} % (fold)
9898
\left| \sprod{x}{y} \right| \le \| x\| \cdot \| y\|
9999
\end{array}\end{math}\\
100100
$\mathbb{K}$ steht für $\mathbb{R}$ und $\mathbb{C}$\\
101-
$\mathbb{I}_n$ ist die $nxn$-Einheitsmatrix.
102-
101+
$\mathbb{I}_n$ ist die $nxn$-Einheitsmatrix. \qquad$e_i$ ist der i-te Einheitsvektor.
103102
\subsection{Matrizen}
104103
% ----------------------------------------------------------------------
105104
Die Matrix $A=(a_{ij}) \in \mathbb K^{m\times n}$ hat $m$ Zeilen mit Index $i$ und $n$ Spalten mit Index $j$.
@@ -128,7 +127,7 @@ \subsubsection{Transponieren}
128127
$(A+B)^\top=A^\top+B^\top$\qquad $(A\cdot B)^\top=B^\top\cdot A^\top$\qquad \\
129128
$(\lambda A)^\top=\lambda A^\top$ \qquad \qquad \qquad $(A^\top)^\top=A$\\
130129
\\
131-
$A\in \mathbb K^{n\times n}$ ist symmetrisch, falls $A=A^\top$\qquad ($\Rightarrow$ diagbar)\\
130+
$A\in \mathbb K^{n\times n}$ ist symmetrisch, falls $A=A^\top$ ($\Rightarrow$ orth. diagbar)\\
132131
$A\in \mathbb K^{n\times n}$ ist schiefsymmetrisch, falls $A=-A^\top$\\
133132
$A\in \mathbb K^{n\times n}$ ist orthogonal (Spalten-/Zeilenvektoren=ONB), falls:\\
134133
$AA^\top=\mathbb{I}_n \quad \Leftrightarrow \quad A^\top=A^{-1} \quad \Leftrightarrow \quad \det A=\pm 1$\\
@@ -154,15 +153,19 @@ \subsubsection{Inverse Matrix von $A\in \mathbb K^{n\times n}$}
154153

155154
\subsubsection{Rang einer Matrix $A\in \mathbb K^{m\times n}$}
156155
{\tiny (N0-Zeilen = Nicht-Null-Zeilen)}\\ \\
156+
\begin{minipage}{\columnwidth}
157157
\textbf{Bringe A auf ZSF} \\
158158
Rang (Zeilrang) $\rang(A)$: Anzahl N0-Zeilen \\
159159
Zeilenraum $\row(A)$: Erzeugnis der Zeilen, $\text{Basis}(\row(A)) = \{\text{ N0-Zeilen }\}$ \\
160160
Kern: $\Kern(A) = \dme{x \in \mathbb K^n}{Ax= 0}$ \\
161161
Dimensionsformel: $\rang(A) + \mathrm{dim}(\Kern(A)) = n$ \\
162+
\end{minipage}
163+
\begin{minipage}{\columnwidth}
162164
\textbf{Bringe A auf Spaltenstufenform (transponieren, ZSF)} \\
163165
Spaltenrang: Anzahl der N0-Spalten\\
164166
Spaltenraum $\col(A)$: Erzeugnis der Spalten, $\text{Basis}(\col(A)) = \{\text{ N0-Spalten }\}$ \\
165167
Bild = Spaltenraum: Erzeugnis der Spalten
168+
\end{minipage}
166169
\subsubsection{Matrixpotenzen}
167170
Gegeben: $A \in\mathbb{R}^{mxn}, x \in\mathbb{R}^n$.\\
168171
Gesucht: Lösung von $A^n$.\\
@@ -195,8 +198,9 @@ \subsubsection{Lineares Gleichungssystem LGS}
195198
\subsubsection{Determinante von $A\in \mathbb K^{n\times n}$: $\det(A)=|A|$}
196199

197200
\begin{itemize}\itemsep0pt
198-
\item $|A|=\sum\limits_{i=1}^n (-1)^{i+j} \cdot a_{ij} \cdot |A_{ij}|$ \qquad Entwicklung n. $j$-ter Spalte
199-
\item $|A|=\sum\limits_{j=1}^n (-1)^{i+j} \cdot a_{ij} \cdot |A_{ij}|$ \qquad Entwicklung n. $i$-ter Zeile
201+
\item $|A|=\sum\limits_{i\text{ oder }j=1}^n (-1)^{i+j} \cdot a_{ij} \cdot |A_{ij}|$ \\
202+
Entwicklung nach $j$-ter Spalte oder $i$-ter Zeile
203+
%\item $|A|=\sum\limits_{j=1}^n (-1)^{i+j} \cdot a_{ij} \cdot |A_{ij}|$ \qquad Entwicklung n. $i$-ter Zeile
200204
\item $\det\begin{pmatrix}A&0\\C&D\end{pmatrix}=\det\begin{pmatrix}A&B\\0&D\end{pmatrix}=\det(A)\cdot\det(D)$
201205
\item $\begin{vmatrix}\lambda_1&&* \\ &\ddots& \\ 0&&\lambda_n \end{vmatrix} = \lambda_1\cdot \ldots\cdot \lambda_n = \begin{vmatrix} \lambda_1&&0 \\ &\ddots& \\ *&&\lambda_n \end{vmatrix}$
202206
\item $A=B \cdot C \quad \Rightarrow \quad |A|=|B| \cdot |C|$
@@ -309,19 +313,34 @@ \subsubsection{Orthogonalität}
309313
\item \textbf{Orthonormalsystem}, wenn $B$ Orthogonalssystem u. $\forall v\in B: \norm{v}=1$
310314
\item \textbf{Orthonormalbasis(ONB)}, wenn $B$ Orthonormalsystem u. Basis von V ist
311315
\end{itemize}
312-
\textbf{Matrix $A$ heißt orthogonal}, wenn $A^\top A = \mathbb{I}_n$
313-
\begin{itemize}\itemsep0pt
316+
Eine \textbf{quadratische Matrix $A$ heißt orthogonal}, wenn $A^\top A = \mathbb{I}_n$
317+
\begin{tabular}{p{0.45\linewidth}p{0.45\linewidth}}
318+
\begin{itemize}
314319
\item $A^{-1}=A^\top $
315320
\item $\det{A}=\pm1$
316321
\item Spalten bilden ONB
317322
\item Zeilen bilden ONB
318-
\item $\norm{Av}=\norm{v}$
319-
\end{itemize}\itemsep0pt
323+
\item $\|Av\|=\|v\|$ % Using \norm for the norm
324+
\end{itemize}
325+
&
326+
\begin{itemize}
327+
\item Drehmatrix mit Drehung um den Ursprung:\ \
328+
$\begin{pmatrix}
329+
\cos(\alpha) & -\sin(\alpha)\\
330+
\sin(\alpha) & \cos(\alpha)
331+
\end{pmatrix}$
332+
\item (Dreh-)Spiegelmatrix mit Spiegelung an der Geraden $y=tan(\frac{\alpha}{2})\cdot x$:\ \
333+
$\begin{pmatrix}
334+
\cos(\alpha) & \sin(\alpha)\\
335+
\sin(\alpha) & -\cos(\alpha)
336+
\end{pmatrix}$
337+
\end{itemize}
338+
\end{tabular}
320339
\textbf{Orthonormalisierungsvefahren einer Basis $\{v_1,\ldots,v_n\}$ nach Gram-Schmidt}
321340
\begin{enumerate}\itemsep0pt
322-
\item $b_1=\frac{v_1}{\|v_1\|}$ \qquad (Vektor mit vielen 0en oder 1en)
323-
\item $b_{2}= \frac{c_2}{\norm{c_2}}$\ \ mit \ \ $c_2=v_2-\frac{\sprod{v_2}{v_1}}{\sprod{v_1}{v_1}}\cdot v_1$
324-
\item $b_{3}= \frac{c_3}{\norm{c_3}}$\ \ mit \ \ $c_3=v_3-\frac{\sprod{v_3}{v_1}}{\sprod{v_1}{v_1}} \cdot v_1-\frac{\sprod{v_3}{c_2}}{\sprod{c_2}{c_2}}\cdot c_2$
341+
\item $b_1=\frac{c_1}{\|c_1\|}$\ \ mit \ \ $c_1=v_1$ \ \ (Vektor mit vielen 0en oder 1en)
342+
\item $b_{2}= \frac{c_2}{\norm{c_2}}$\ \ mit \ \ $c_2=v_2-\frac{\sprod{v_2}{c_1}}{\sprod{c_1}{c_1}}\cdot c_1$
343+
\item $b_{3}= \frac{c_3}{\norm{c_3}}$\ \ mit \ \ $c_3=v_3-\frac{\sprod{v_3}{c_1}}{\sprod{c_1}{c_1}} \cdot c_1-\frac{\sprod{v_3}{c_2}}{\sprod{c_2}{c_2}}\cdot c_2$
325344
\end{enumerate}
326345
\textbf{Erweitern einer ONB von $V$ auf eine ONB des $\mathbb{R}^n$}
327346
\begin{enumerate}\itemsep0pt
@@ -481,42 +500,45 @@ \subsection{QR-Zerlegung}
481500
$A = QR$, wobei $Q$ orthogonal und R oben dreieckig.\\
482501
\textbf{Vorgehen}
483502
\begin{itemize}\itemsep0pt
484-
\item $Q$ berechnen durch Gram-Schmidt mit den Spalten von $A$, beginnend bei der ersten
485-
\item Die Koeffizienten von $R$ ergeben sich durch Umstellen der jeweiligen Gram-Schmidt Gleichungen auf die Spalten von $A$
503+
\item $Q$ berechnen durch Gram-Schmidt mit den Spalten von $A$, \textbf{beginnend bei der ersten}
504+
\item Die Koeffizienten von $R$ ergeben sich aus den Gram-Schmidt Gleichungen wie folgt: $r_{i,i}=||c_i||^2$ und $r_{i,j}=\frac{\langle v_j,c_i \rangle}{r_{i,i}}$
486505
\item Alternativ gilt: $R = Q^TA$
487506
\end{itemize}
488507

489508
\subsection{Kleinstes-Quadrate-Problem}
490509
Für $Ax = b$ lautet die
491510
\textbf{Normalengleichung}
492-
$A^TAx = A^Tb$\\
511+
$A^TAx^* = A^Tb$\\
512+
\\
513+
\textbf{Lösen} durch Gauß oder Umstellen: $x^* = (A^TA)^{-1}A^Tb$\\
514+
Für $A=QR$ lautet die Lösung $x^* = R^{-1}Q^Tb$\\
493515
$\Rightarrow$ optimale Lösung mit minimalem quadratischen Fehler (existiert immer).
494516

495517
\subsection{Singulärwertzerlegung}
496-
Bei der Singulärwertzerlegung wird eine beliebige Matrix $A\in \mathbb{R}^{m\times n}$ als Produkt dreier Matrizen $U$, $S$ und $V$ geschrieben
518+
Bei der Singulärwertzerlegung wird eine beliebige Matrix $A\in \mathbb{R}^{m\times n}$ als Produkt dreier Matrizen $V$, $\Sigma$ und $W$ geschrieben
497519
\begin{equation*}
498-
A=USV^\top
520+
A=V\Sigma W^\top
499521
\end{equation*}
500-
mit $U\in \mathbb{R}^{m\times m}$, $S\in \mathbb{R}^{m\times n}$ und $V\in \mathbb{R}^{n\times n}$.\\
501-
$U$ und $V$ sind orthogonal, $S$ ist eine Diagonalmatrix.
522+
mit $V\in \mathbb{R}^{m\times m}$, $\Sigma\in \mathbb{R}^{m\times n}$ und $W\in \mathbb{R}^{n\times n}$.\\
523+
$V$ und $W$ sind orthogonal, $\Sigma$ ist eine Diagonalmatrix.
502524
\subsubsection{Rezept: Singulärwertzerlegung}
503525
Gegeben: $A\in \mathbb{R}^{m\times n}$
504526
\begin{enumerate}\itemsep0pt
505-
\item Bestimme alle Eigenwerte $\lambda_j$ und Eigenvektoren $v_j$ der Matrix $A^\top A\in \mathbb{R}^{n\times n}$ und ordne sie \\ $\lambda_1\ge\lambda_2\ge \dots \ge \lambda_r>\lambda_{r+1}=\dots=\lambda_n=0$ mit $r\le n$
506-
\item Bestimme eine ONB des $\mathbb{R}^n$ aus den Eigenvektoren $v_j$ und erhalte $V=\begin{pmatrix}
507-
v_1 &\dots & v_n
527+
\item Bestimme alle Eigenwerte $\lambda_j$ und Eigenvektoren $w_j$ der Matrix $A^\top A\in \mathbb{R}^{n\times n}$ und ordne sie \\ $\lambda_1\ge\lambda_2\ge \dots \ge \lambda_r>\lambda_{r+1}=\dots=\lambda_n=0$ mit $r\le n$
528+
\item Bestimme eine ONB des $\mathbb{R}^n$ aus den Eigenvektoren $w_j$ und erhalte $W=\begin{pmatrix}
529+
w_1 &\dots & w_n
508530
\end{pmatrix} \in \mathbb{R}^{n\times n}$
509531
\item Die Singulärwerte sind $\sigma_j=\sqrt{\lambda_j}$ \qquad $j=1,\dots,\min\{m,n\}$
510532
\begin{equation*}
511-
S=\begin{pmatrix}
533+
\Sigma=\begin{pmatrix}
512534
\sigma_1 & & & 0 & \dots & 0\\
513535
& \ddots & & \vdots & & \vdots\\
514536
& & \sigma_m & 0 & \dots & 0\\
515537
\end{pmatrix} \in \mathbb{R}^{m\times n}
516538
\qquad m<n
517539
\end{equation*}
518540
\begin{equation*}
519-
S=\begin{pmatrix}
541+
\Sigma=\begin{pmatrix}
520542
\sigma_1 & & \\
521543
& \ddots & \\
522544
& & \sigma_n \\
@@ -526,32 +548,90 @@ \subsubsection{Rezept: Singulärwertzerlegung}
526548
\end{pmatrix} \in \mathbb{R}^{m\times n}
527549
\qquad m>n
528550
\end{equation*}
529-
\item Bestimme $u_1,\dots,u_r$ aus $u_i=\frac{1}{\sigma_j}Av_j$ für alle $j=1,\dots,r$ (alle $\sigma_j\ne 0$)
530-
\item Falls $r<m$ ergänze $u_1,\dots,u_r$ zu einer ONB, bzw. zu $U=\begin{pmatrix}
531-
u_1 & \dots & u_m
551+
\item Bestimme $v_1,\dots,v_r$ aus $v_i=\frac{1}{\sigma_j}Aw_j$ für alle $j=1,\dots,r$ (alle $\sigma_j\ne 0$)
552+
\item Falls $r<m$ ergänze $v_1,\dots,v_r$ zu einer ONB, bzw. zu $V=\begin{pmatrix}
553+
v_1 & \dots & v_m
532554
\end{pmatrix}$
533555
orthogonal.
534-
\item $A=USV^\top$
556+
\item $A=V\Sigma W^\top$
535557
\end{enumerate}
536-
537-
\subsection{Lineare Differentialgleichungen}
558+
\subsubsection{Kompression}
559+
\begin{itemize}\itemsep0pt
560+
\item Rang k Matrix $A_{(k)}=V\Sigma_{(k)}W^\top$. Die Matrix $\Sigma_{(k)}$ enthält nur die Singulärwerte $\sigma_1, ...,\sigma_k$. ($\sigma_{k+1}, ...,\sigma_{n/m}$ mit 0 ersetzen).\\
561+
\item Frobenius-Norm $|| A ||_F = \sqrt[]{\sum_{i=1}^{m} \sum_{j=1}^{m} {a^{2}_{i,j}}}$\\
562+
\item Es gilt: $|| A-A_{(k)} ||_F \leq || A-B ||_F$ mit $B$ als eine beliebige Matrix des $\mathbb{R}^{m\times n}$ mit $rang(B) \leq rang(A_{(k)}) = k$
563+
\item Speicheraufwand einer Matrix $A\in \mathbb{R}^{m\times n}$: $rang(A) \cdot (m+n)$
564+
\end{itemize}
565+
\subsection{Lineare Differentialgleichungen und Rekursive Folgen}
538566
\subsubsection{Lösen einer linearen Differentialgleichung}
539567
Gegeben: $y'(t) = \lambda y(t)$, mit $y(0) = c$\\
540568
Lösung: $y(t) = ce^{\lambda t}$ mit $c \in \mathbb R$\\
541569
\subsubsection{Lösen eines Systems linearer Differentialgleichungen}
542-
Gegeben: $y'(t) = Ay$, mit $y_1(0) = c_1 , ..., y_n(0) = c_n$\\
543-
Wobei $A \in \mathbb R^{nxn}, c_1, ..., c_n \in \mathbb R, y,y'\in\mathbb R^n$.\\
570+
Gegeben: $y'(t) = Ay$, mit $y_1(0) , ..., y_n(0)$\\
571+
Wobei $A \in \mathbb R^{nxn}, y_1(0) , ..., y_n(0) \in \mathbb R, y,y'\in\mathbb R^n$.\\
544572
Lösung:\\
545573
\begin{enumerate}
546574
\item Eigenwerte und Eigenvektoren von $A$ bestimmen.
547575
\item $\begin{pmatrix}
548-
x_1(t)\\
576+
y_1(t)\\
549577
\vdots\\
550-
x_n(t) \\
578+
y_n(t) \\
551579
\end{pmatrix}
552580
= c_1e^{\lambda_1t}v_1+...+c_ne^{\lambda_nt}v_n$.
553581
\item Anfangswerte einsetzen und Werte für $c_1$, bis $c_n$ bestimmen.
554582
\end{enumerate}
583+
\subsubsection{Rekursive Folgen}
584+
Gegeben: $x_{n+1} = {\alpha} x_n + {\beta} x_{n-1}$, Anfangswerte: $x_0$, $x_1$\\
585+
Lösung:\\
586+
\begin{enumerate}
587+
\item Matrix A bestimmen.
588+
\begin{equation*}
589+
A=\begin{pmatrix}
590+
{\alpha}&{\beta}\\
591+
1&0
592+
\end{pmatrix}
593+
\end{equation*}
594+
\item LGS aufstellen.
595+
\begin{equation*}
596+
\begin{pmatrix}
597+
x_{n+1}\\
598+
x_{n}\end{pmatrix}=\begin{pmatrix}
599+
{\alpha}&{\beta}\\
600+
1&0
601+
\end{pmatrix}
602+
\begin{pmatrix}
603+
x_{n}\\
604+
x_{n-1}\end{pmatrix}
605+
\end{equation*}
606+
\item Diagonalisieren: $A=SDS^{-1}$
607+
\item Anfangswerte $x_0, x_1$ einsetzen und $x_n$ berechnen
608+
\begin{equation*}
609+
\begin{pmatrix}
610+
x_{n+1}\\
611+
x_{n}\end{pmatrix}=SD^{n}S^{-1}
612+
\begin{pmatrix}
613+
x_{1}\\
614+
x_{0}\end{pmatrix}
615+
\end{equation*}
616+
\end{enumerate}
617+
\subsection{Definitheit und Quadratische Funktionen}
618+
\subsubsection{Definitheit}
619+
Eine symmetrische Matrix $A \in\mathbb R^{n\times n}$ mit den Eigenwerten (EW) $\lambda_1, ..., \lambda_n$ heißt:
620+
\begin{itemize}\itemsep0pt
621+
\item positiv definit: EW positiv $<=> v^{\top}Av> 0,\quad\forall v \in\mathbb R^n\backslash \{0\}$
622+
\item negativ definit: EW negativ $<=> v^{\top}Av< 0,\quad\forall v \in\mathbb R^n\backslash \{0\}$
623+
\item positiv semidefinit: EW $\geq 0 <=> v^{\top}Av\geq 0,\quad\forall v \in\mathbb R^n$
624+
\item negativ semidefinit: EW $\leq 0 <=> v^{\top}Av\leq 0,\quad\forall v \in\mathbb R^n$
625+
\end{itemize}
626+
\subsubsection{Quadratische Funktionen}
627+
\textbf{Form:} $f(x)=x^{\top}Ax+b^{\top}x+c = \langle x,Ax \rangle + \langle b,x \rangle + c$\\
628+
Berechnen von Extrempunkten:
629+
\begin{itemize}\itemsep0pt
630+
\item positiv definit: Minimum bei $x^{*}=-\frac{1}{2}A^{-1}b$
631+
\item negativ definit: Maximum bei $x^{*}=-\frac{1}{2}A^{-1}b$
632+
\item positiv/negativ semidefinit: Existenz von Extremum hängt von Lösbarkeit des LGS $2Ax=-b$ ab. (nicht eindeutig!)
633+
\end{itemize}
634+
555635
\end{multicols*}
556636
% Ende der Spalten
557637

0 commit comments

Comments
 (0)