0% found this document useful (0 votes)
14 views16 pages

HW2 110060040 蔡奎元

1) The document discusses solving systems of linear equations using Gaussian elimination. 2) It shows that the solutions to a homogeneous system form a subspace, by proving three properties. 3) The document solves an example system of equations to find a particular solution and describes how the general solution can be written in terms of the particular solution and the nullspace.

Uploaded by

蔡奎元
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views16 pages

HW2 110060040 蔡奎元

1) The document discusses solving systems of linear equations using Gaussian elimination. 2) It shows that the solutions to a homogeneous system form a subspace, by proving three properties. 3) The document solves an example system of equations to find a particular solution and describes how the general solution can be written in terms of the particular solution and the nullspace.

Uploaded by

蔡奎元
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

縁燻代灘 HW 2

Linear Alqebra

"

感的

òii㦝
.io
ˊ

ˋ
(a) 3 unknow
thereare

ns.yetonlyzequationsithereareinfinitesolutionstolxi.X2.X 3)

(b)
thecorrespondinghomogeneoussystemis

Iiiiǒl
Gaussian
-
[ iii)
-

1 |
Elimination 1 0
1比 A =
> = R
0 1 i

wewanttosolve Aiō Rtō


訓贓

※xz -_-
我3

州 絒獸
showthattnesolutionsofxformasubspaceinR.wellhave.to
To

show 3 things
:

① 0 E i

② ifu.VE 女 utiei

③ ifū Ei axūtx (aE R)

|別

i = ㄨ3 x takex3 0 女 = 0

oeō


letnikx
闹 ti ; i =
a
闹 Ei

ūtū =
( ktl )
|別 Ei
𠶜


letnikx ti

axū = akx
闹 Ei

Nowthatwéveproven ① 、
② ˋ

thesolutionsofxformasubspace.in 成 ⼀

!別
(C)

KHEK x_x
}
since
|別 alonespansthesubspacekt

蹄| formsabasisfor KH .

(d)
letssolvethissystemofequations :

Iiii 别 [ 7
7
1 2 1
"" >
0 -
3 2
-
-
11

-
Iòiiililòiiiil
1哈 憫瑜
Xnitix
1
3

xij -53
" "

州絒 j.is
0 t
代跳悯
X3


special 下

v1 訂 isaspecialsolutionr

(e) thesystemcanbewrittenas :

[ 囐岡
1 2 1

Tsletthisbevector 5
ninmi
letthisbe letthisbe
matrix A vectorx

suppose

thesetkisthegeneralsolutionofAx-5ithevectortisaspe.ci
alsolutionof Ai -5 ;
andtheset KH istnegeneralsolutionof Ai ⼆
0

i.e-KHistnenulspaceofA.rci?)considerthesetk-i,A(k-t)=Ak-Ai=5
(

-5 =
0

K-iisthenullspaceofA.in (ii)

By (i). (ii) KÚ ⼆
KH

k ⼆ Ūt KH

thecosetttKHisthegeneralsolutionofthesyste.mx
(f) of HWI weleamedthat
Inproblem 4.la ) ,
:

If Wisasubspaceofthevectorspace V
and ŪEV ,

Ūt Wisasubspaceof V ÚEW

Inthisproblem ,

wherei.KHarebothinpiHowever.VE/KH,therefore,K=i+KHdoes
KEŪ t KH ,

notformasubspaceinpi .
" "
(1) showthat spanl.SN 52) Espan (六) tspanl.SI
letxtspanl.SN 品)

T ˋ
⼤ … …
Xk ESiandJI.ir …
.IE 52

suchthattaitait-tarntbiitbzizt.at5 mkx
let ū ⼆ a.it ait-tanxtspanl.si )
F = biiktltbziktzictbmkx Espan ( 5 ) 2

iūtū Espan (G) +


spanlsz )

sina.IE span 你 V52) implies


XE span (G) tspan ( S ) 2

spanlS.US 2) ≤ spanlS.lt spanl.SI

" "
(2) showthat span 你 ) tspanl.SI Espanl.SN 52)

let.IE span (G) +


spanlsz )

theremustexistūtspan (G) and Ūtspan (品 )

suchthat ieūtt
sinceutspanl.sn
T.i.i.IE Si ,

suchthatu-axtal-tarnsinceitspanl.SI
Ixn.in …
.in E 52
,
,

suchthat Ū ⼆ bihtbzizt …
tbnnx

iūti

GIikfieitbiitbzit-tbn-risincei.E.o.o.it

,
Si

ii. …
在 ESIUS 2
ǎnanuneynn …

sincetkti.cz … ˋ
XE Sz
v
Fktl 在2
ˋ
… ˋ
⼤ E Si V52

bii.biz … …
bn-kXEspanlS.US ) …
(ii)

By (i). (ii) IE spanlS.US 2)

sina.IE span (G)


spanlSzlimplie.si
+


Espan V52)

spanlS.lt spanlSDEspanlS.US 2)
(3) sinceweieshownthat

(I) spanlS.US 2) EspanlS.lt spanl.SI


(2) spanlS.lt spanlSDEspanlS.US 2)

spanlS.US 2 ) =
span 你 ) tspan ( 5 2) *

Mi
-

Mii M 1 2 MB …

0 Mzz
M23-Mznlet.ME 0 0 M 33 …
M3n

:
ii_OOO.inMnnnxn
i :

andlet.MX:0
ert e re
ee he e re
ee

Mii M 1 2 MB …
Mm Xl 0

0 M 22 M 23
-
Mzn Xz 0

I. e. 0 0 M 33 …
M3n X3 = 0

i : : i i i :

OOO.in Mnn Xn 0
_ -

nxn
- -
nxl - -
nxl

M i ō
in.io
⼀ -

ii

Mn
○ Mzz Mzn 0

t.n.tl/nM3n=Oi
Xi 0 + Xz 0

i : i

tn.O_sinceMii.Mzzn.in
0 0
- _ _ _


Mnn areallnomzero
ee e re e re
ee

x 0

Xz 0

I = X3 = 0

i :

N_n _
0
_

the.co/umnsofMarelinearlyindependent. ☒
(a)
Subspace Basis Dimension

W.nwzlnnii.in } k
Iet
Wi t.nin.n.i.ii.in }
mtkwzfn.nin.li , iii) ptk

Wewanttoprovethat :

dimlwit Wz ) =
dmlWDtdmlwzl-dmlw.nu )
= ( mtk ) + ( ptk ) -

=
ktmtp

Nowconsidertheset

T =
{ n.n.in , t.ii-iim.i.int }

Ifwecanshowthat Tformsabasisfor Wit Wz ,

thenwecanobtainthatdmlw.tw ) 2
=
ktmtp
① showthatthevectors.in Tspan Wit Wz

IEWitwzcanbewrittenasx-p.it
Anyvector

tpkǘktqūtqit
PŪ + … …

tqin
+ r.it rtzt …

trp Ūp ,

whichisalinearcombinationofthevectors.in
Ti

thevectors.in Tspan Witwz .

② showthatthevectorsin Tareindependent

letawitahi-takhtbiitbnt.at bmtm
+ aūtaūtntcp Ūp ⼆ Ō

aūtazǖztiakhtbiǖtbnt …
tbmtm
-_- cnn.ci - . -

Cpn … … …
(☆)

cnn.ci - . -

cpopEW2.n.li )

( smce-cni-aui-cpnisalinearcombinationjofci.ci
EW2-cni.ci
… …

Ūp

- . -

GUEWi.no (ii)

( smce-cni-aui-cpnisalinearcombinationjofni.no
viii. .in EW, …
By (i). (ii)

cni-ai-n-GWTEW.in Wz
cni-cz.ci
cpopcanbewrittenasalinearcombinationofn.in
-
-,,-
-

Ūk

whichisabasisofhinWz.let-cni-cz.ci
… …

-
-,,-

cpop-dntdtt.it dkūn
GŪ taito.it Cp Ūptdūtd! + …
tdkūk ⼆ Ō

sincecici-ci.in … ⼀
Ūkforma
basisfor Wz

theyarelinearlyindependent
a =
Cz =,=
G =
d = dz =_= dk 0

Putting Czt G 0 backin (A)


givesusawitaht-takhtbiitbnt.at
G … = =

bmtm ⼆
Ō

Sinceüi …
Úriǖü …
Ūmforma
basisfor Wl

theyarelinearlyindependent .

a = a = …
= a = b, = bz =_= bm = 0
Returningtoourinitialassumption :

aūtazǖztiakhtbiǖtbnt …
tbmtm
ta Ūtaūt …

tcp Úp ⼆ Ō

Sincewediscoveredthat
G =
a = …
=
G =
bi =
bz = _ = bm =
G = G = … =
G =
0

ün …
n.it …
不 ui.ci …

.ci

arelinearlyindependent.BY
① & ② ,

t.i -i m.w.i .iarelinearlyindependentandspanW.tw


thevectorsvi.n.in ,

T-fn.n.in , t.ii-iin.ni.int }
formabasisfor Wit Wz

dmlwitwz ) = k+ m +
p
=
dmlw.lt dmlwz) -

dmlwnwj
(b) Given V = Wit Wz ,

inn-n.wnnn-ts.gs
pf " "

Visthedirectsumof Wiand Wz

Visthedirectsumof Wiand Wz
V = W, + Wz and Win Wz ⼆
∅ ( Bydef )

In (a) ,
welearnedthat

dim ( Witwz ) = dim (vi) tdimlWD-dimlw.nu )


dim (V) =
dimlwiltdim (vi) -
dim ( ∅ )
=
dimlW.lt dimlwz)
*

nnnnnn-rt-D.gs/iVisthedirectsumofWiandW2
f "

"

"
dimlWDtdimlwy.o.by given
condition
dim (V) =
dimwtwy
idim (vi) tdimlwz )
-

dim ( Winwz ) …

.by (a)
dimlwnwz ) = 0

Wnwz =

Since V = Wit Wz and Win Wz ∅

Visthedirectsumof Wiand W2.*

You might also like