17
Matrix representation Notation : An ordered basis for a finite - dimensional vector space V is a basis for V endowed with a specific order ( es IR ' Ello ) , 1913¥ E (9) ill ) } as " ordered I I p , P2 basis ) Definition ' Let V be a finite - dimensional vector space and p={ vii. ii. , , In ) be an ordered basis for V Then , tix EV , I ! a. , az , - . , an EF sit . I = II , ai Ii The coordinate vector of I relative to p , denoted as EX ]p , is the coluqrfn , vector Fine - - ( ha EF 's

P2 basis - Chinese University of Hong Kong · Matrix representation Notation: An ordered basis for a finite-dimensionalvector space V is a basis for V endowed with a specific order

  • Upload
    others

  • View
    6

  • Download
    0

Embed Size (px)

Citation preview

  • Matrix representation

    Notation : An ordered basis for a finite - dimensionalvector space V

    is a basis for V endowed with a specificorder .

    ( es . IR'

    Ello ) , 1913¥E (9) ill )} as" orderedI Ip , P2 basis )

    Definition ' Let V be a finite - dimensional vector space and

    p={ vii. ii. , . . . ,In ) be an ordered basis for V .

    Then , tix EV , I ! a. , az ,- .

    ,an EF sit . I = II

    ,

    ai Ii.

    The coordinate vector of I relative to p , denoted asEX]p ,

    is the coluqrfn,vector Fine

    --

    ( ha EF's

  • Remark " Define a map V → Fn . This map islinear

    Its CI3p( HW - [ a

    'IyF3p = a -ii3ptCY3p )

    Now , suppose V andW are finite - dimensional vector spaces

    with ordered bases p - { iii. Iz , . . ,In } and 8=1 WT , . . . , Tim }

    ( for V ) ( for W )respectively .

    Let T - . V -1W be alinear transformation .

    Then for each IEJ Eh ,I aijef csis m such

    that

    T ( rij ) = ,÷,

    aijwi for a En ,A

    w

  • Definition : with this notation as above , we call the matrix

    A ( aij ) # em thematrix representation

    I s n

    of T in the ordered bases Band I

    ,and

    denoted it as A = I Ttp .If V = W and p

    -- 8

    ,then we simply

    write [T3§X C TIP

    T ( iij ) = it,

    aijwi for a En ,a

    w

  • T ( Tj ) =,

    aijwi for i En ,

    . i:o÷:÷::÷④÷÷÷÷÷:÷:* .Hi:D , cab Fiends

  • p= { Ii , Ja , . . . , Tin } for V

    j = { WT , Ja ,. .

    .

    ,I'm } for W

    :÷::*ina .-n

  • Examples : Fn

    • Let AEMmxnlH.LA: F " → Fm defined by = LA ( ¥)Et AI

    '

    Let p,

    and 8 be thestandard bases for F

    "

    and Fm resp .

    { (b:o) , ) . . } nth col of A

    first Col . of A

    I Latif . . . cafeAf !) - - firstof = ftp.n.ntjo = A

  • • For T - - Pnllk ) → Pm, LIR) defined as Tffcxi ) = fix , .

    Let p= { I , X , X ? ..

    .

    ,X

    " } be an ordered basis forPnllr )

    Let D= { I , X ,X

    '

    ,-

    ..

    ,X

    " } be an ordered basis forPn-11112)

    rn : repin. ."

    ÷

    .

    .. .

  • Lecture 10:

    Recall : . Coordinate representation ofTi EV w - r - t . B

    ,

    Write Ii = a. I i tazvzt . . . tannin { it , , . . . , In }

    I E I -

    [ Its =/ !!) e Fnordered

    • Matrix representation of a T: V → W

    p 8

    TEE. ,! . .in }

    '

    II . , . . . , m3

    ITI -- ftp.t, . . . . ITme Mmxn

  • Example : T : Mzxzllk ) → Mzxz L IR ) defined by :

    TCA ) EetAT t 2AT

    tram pose

    P = { ('

    o: ) , ( :! ) , ( o, ;) , ( °o°, )

    3 - ordered basis

    a m-

    - K::) .li : ) .co . :) . c ::Xi Tip

    -

    -

    ( ! ! ! ! )

  • Example : T : Pz L IR ) → Mzxz CIR ) defined by :

    Tcf ) Eet (Sto ) fu ) )O f ' C o )Consider ordered basis =

    p = { I , × , ×' } for PLUR )

    a- a ::) . c :: , .ci :3 . c ::B

    Tim - - El :: ) . c :'

    die ::D

    :-C Ttp =/ ! ! 1) E Max ,

  • Composition of linear transformations and matrix multiplication

    Thin : Let V and W be twovector spaces

    over the same field 't.

    And let T : Vs W and U' - W -32 be linear

    .my -( it Then the composition UT : V → 2 ① I ⑥→u⑦is linear .a p 8

    ( ii ) If V , w , Zhaveordered bases L

    , p , 8

    rnespeclivmely,

    P

    then : [ UT ] ! = [ U ] ! [ T ) ! E Mmxn

    a ←matrix multiplication .Mpxn Mpxm

  • ( it Let I'

    , 5'

    EV and AEF.

    Then :

    UT Catty ) = U ( atx I -1715'

    )) = a UT Cx ) t U Tty )

    a

    '

    . UT is linear .

    ( ii ) Suppose a-

    - { I , .ir , . . . , In }

    p = I hi , , ii a , . . ,I'm }

    8 = { E.,

    Iz,

    ..

    ,Ep ]

    [ UH = ftp.n-tjjlaih) :{inemmeans : Uco 's , =

    ,

    air Ii

    Alka- k isham=

    I l: IApkT

    '

    heth

  • [ TIE = B Eet l bkj ) :mnmeans Thijs z!bkjTika

    Mmxnlf ) for I Ey

    Then : UT ( Tj ) = Ul ¥,

    bkjwk )= ¥

    ,

    beg. Utwk )

    =E?bkj⇐aiaEi) - - ¥, '

    i

    TC is j I - entry of

    So,

    CUT ] ! = AB = [ U ] ! I TIP AB

  • × pExample : Consider : U : penury → Pm C IR )

    defined by i

    U Cfcxi )Eet ft × ,Consider T :pmfcuz, →paindefined by :Tcfcx ) ) Etf! ftldtLet d and p be standard ordered bases for Pn CIR ) and Put UR )

    respectively,

    in :p:÷÷÷.

    .

    :. " " I÷÷:÷ .

    So,

    [ UTI,

    = CUTIE = Idgaf.li?--EIpn..umIp

  • Corollary : Let V and W be finite -dimensional Vector spaces

    with ordered basis p and8 respectively .

    Let T : V → W be linear . Then : for anyI EV

    ,we have :

    [Tignish

    = CTI ; e p-

    I Matrixmultiplication

    Lin.

    Transf.

    Proof : Fix rt e V and consider twolinear transformations :

    j& p

    f : F → V g: F → W

    defined bydefined by

    flags= art EV g la ) = a Thi )

    E W

    f and g are linear transformations .Also , 9 = To -5 .

  • Let d = { I } be the standard orderedbasis for F

    .

    [ Thing = C gaily = C g) I= IT ] } I ftp.CTJpcfc/Dp

    To't = E Ttp E utpx p 8

    FE, VI. wLEI }

    g8 -¥f

    F - w

    Egg ! -_() - - (&'D!

  • Example : T : Mzxz I IR ) → Muz C IR) defined by :

    TCA ) Eef AT -12A .

    is -- El ::) , c :: , .ci : , ,c :: , }

    uh - - f ! ! ! ! )

    iii.'

    i÷ : : : in it:L:c :"