Matrix multiplication — definition, associativity, non-commutativity: Foundations

4.5.7 · D1 · Maths › Linear Algebra (Full) › Matrix multiplication — definition, associativity, non-commu

Parent note padhne se pehle, tumhe kuch symbols ko achhi tarah samajhna hoga. Yeh page unhe ek-ek karke zero se build karta hai: pehle plain words mein, phir ek picture, phir kyun yeh topic unke bina chal nahi sakta. Hum dependency order mein jaate hain — har idea sirf upar wale ideas pe lean karta hai.

1. Ek box mein numbers — matrix asal mein hoti kya hai

Plain words mein: ek matrix bas ek numbers ki table hai, kisi spreadsheet se zyada mysterious nahi. Jaadu yeh hai ki hum baad mein iske saath kya karte hain.

Neeche ki picture mein wahi grid hai jisme rows aur columns colour se dikhaye gaye hain — is layout ko apni aankhon mein burn kar lo, kyunki har baad ka rule "row pakdo, column pakdo" wala hai.

Kyun topic ko iska zaroorat hai: poora parent note in grids ko combine karne ke baare mein hai. Agar tum "row 2" aur "column 1" ko turant point nahi kar sakte, toh row·column rule geography ki jagah magic lagegi.

2. Andar ek number ko naam dena — subscript $A_{ij}$

Matrix $A = (1324)$ ke liye:

$A_{11} = 1$ (row 1, column 1),
$A_{12} = 2$ (row 1, column 2),
$A_{21} = 3$ (row 2, column 1),
$A_{22} = 4$ (row 2, column 2).

Kyun topic ko iska zaroorat hai: master formula $C_{ij} = \sum_{k} A_{ik} B_{k j}$ poori tarah inhi addresses se bana hai. Agar $A_{ij}$ fuzzy hai, toh woh formula unreadable hai.

3. Do shape numbers — $m \times n$

$(142536)$ hai $2 \times 3$ : do rows, teen columns.
Ek single column $(57)$ hai $2 \times 1$ .

Kyun topic ko iska zaroorat hai: shape mismatch #1 reason hai ki product exist nahi karta, aur unequal shapes reason 1 hai jo parent non-commutativity ke liye deta hai ( $A B$ ho sakta hai $2 \times 2$ jabki $B A$ ho $3 \times 3$ ).

4. Ek vector — woh cheez jis par matrix act karti hai

Yahan $x_{1}$ aur $x_{2}$ vector ke components hain — uske individual coordinate numbers. Subscript ab sirf list mein position hai (sirf ek column hai, toh doosre index ki zaroorat nahi).

Kyun topic ko iska zaroorat hai: parent define karta hai product ko yeh demand karte hue ki $(A B) x = A (B x)$ har vector $x$ ke liye. $x$ ko ek grabbable arrow jaane bina, "matrices machines hain jo vectors ko move karti hain" mein kuch move karne wala hi nahi hoga. Poori "machine" story ke liye dekho Linear Transformations.

5. Dot product — do lists se ek number

Practice mein: $(1, 0, 2) \cdot (4, 0, 1) = 1 \cdot 4 + 0 \cdot 0 + 2 \cdot 1 = 6$ .

Kyun topic ko iska zaroorat hai: parent ka headline hai " $(A B)_{ij} =$ row $i$ of $A$ dotted with column $j$ of $B$ ." Har product ki har single entry ek dot product hai. Geometric meaning (length aur angle) ke liye dekho Dot Product.

6. Summation sign $\sum$ — "inhe add karo" ka shorthand

Plain words mein: $\sum$ bas ek loop hai jo cheezein total karta hai. Neeche wala letter ( $k$ yahan) ek throwaway counter hai — woh sirf andar appear karta hai aur answer se gaayab ho jaata hai. Note karo yeh literally §5 ka dot product hai, " $+ \dots +$ " ki jagah loop ke saath likha gaya.

Kyun topic ko iska zaroorat hai: poori definition, associativity proof, aur double sums $\sum_{k} \sum_{j}$ sab $\sum$ mein likhe hain. Yahi woh language hai jisme poora page bolta hai.

7. Special matrices jo milenge

8. Symbols jo baad mein aate hain (taaki kuch surprise na kare)

Abhi inhe compute karne ki zaroorat nahi — bas symbols ko tab pehchanlo jab parent inhe use kare. Powers jaise $A^{2} = AA$ aur diagonalization Matrix Powers and Diagonalization mein hain.

Yeh topic ko kaise feed karte hain

Ise upar se neeche padho: plain grid tumhe addresses aur sizes deta hai; addresses + sizes + sum sign dot product banate hain; dot product plus "ek matrix ek vector ko move karti hai" product rule banata hai; aur product rule do bade theorems mein branch karta hai — associativity aur non-commutativity — parent topic ke.

Equipment checklist

Dahini taraf cover karo aur khud test karo — agar koi bhi jawab fuzzy hai, toh us section ko dobara padho.

A_{ij}

mein, kaun sa subscript row hai?

Pehla wala,

i

; doosra

j

column hai (Row phir Column).

A

hai

3 \times 2

" tumhe kya bataata hai?

A

mein 3 rows aur 2 columns hain (rows pehle, columns baad mein).

(2, 1, 0) \cdot (3, 4, 5)

calculate karo.

2 \cdot 3 + 1 \cdot 4 + 0 \cdot 5 = 10

— matching entries multiply karo, add karo.

\sum_{k = 1}^{2} A_{ik} B_{k j}

kya expand hoga?

A_{i 1} B_{1 j} + A_{i 2} B_{2 j}

—

k

1, 2

pe loop karo aur add karo.

Ek vector geometrically kya hota hai?

Numbers ka ek column = origin se us point tak ek arrow jiske coordinates wahi numbers hain.

A

(size

m \times n

) ko

B

se multiply karne ke liye kaun se dimensions match karni chahiye?

Inner wale:

B

n \times p

hona zaroori hai; result hoga

m \times p

Identity matrix

I

kisi bhi vector ke saath kya karta hai?

Kuch nahi —

I x = x

; yeh do-nothing machine hai.

Commutator

[A, B]

kis ke barabar hota hai, aur zero ka kya matlab hai?

[A, B] = A B - B A

; yeh

0

hota hai bilkul tab jab

A

aur

B

commute karte hain.

Kyun

I

ko commuting ke liye reference kaha jaata hai?

Har matrix

I

ke saath commute karta hai (

A I = I A = A

), toh yeh baseline "hamesha-commute-karta-hai" case hai.

Connections

Matrix multiplication — definition, associativity, non-commutativity — woh parent jiske liye yeh page tumhe tayaar karta hai.
Linear Transformations — "matrix = machine that moves vectors" picture poori tarah.
Dot Product — woh multiply-and-add operation jo har entry use karta hai.
Identity and Inverse Matrices — $I$ , $A^{- 1}$ , aur kyun order reverse hota hai.
Transpose — flip symbol $A^{⊤}$ .
Determinants — stretch-factor number $det A$ .
Matrix Powers and Diagonalization — powers $A^{2}$ aur commuting matrices.