Foundations — Orthogonal matrices — properties, det = ±1

4.5.37 · D1 · Maths › Linear Algebra (Full) › Orthogonal matrices — properties, det = ±1

Parent note ki ek bhi line padhne se pehle, tumhe har ek notation ko kamana hoga. Hum unhe ek-ek karke build karte hain, ek doosre ke upar, bilkul zero se. Neeche kuch bhi yeh nahi maanta ki tumne pehle matrices dekhi hain.

1. Ek vector — ek arrow jiska ek address hai

Picture: socho ek treasure map instruction — "3 kadam east, 4 kadam north". Woh instruction hi vector $\left(\begin{array}{c}3\\ 4\end{array}\right)$ hai. Arrow woh seedha shortcut hai jahaan se tum shuru hue the wahan se treasure tak.

Topic ko iske zaroorat kyun hai: orthogonal matrices aise machines hain jo vectors ko move karte hain. "Move karna" ke baare mein baat karne ke liye, humein pehle us cheez ka naam chahiye jo move ho rahi hai.

2. Length — arrow kitna lamba hai, likha jaata hai $\Vert x\Vert$

Squares ka square root kyun? Us right triangle ko dekho jo arrow horizontal aur vertical ke saath banata hai. Uske do legs $x_1$ aur $x_2$ hain; arrow hypotenuse hai. Pythagoras — "lamba side squared barabar legs ke squares ka sum" — humein seedha formula de deta hai. Square root squaring ko undo karta hai taaki hume actual distance mile, na ki distance-squared.

Topic ko iske zaroorat kyun hai: orthogonal matrices ka poora point yeh hai ki woh length preserve karte hain ($\Vert Qx\Vert =\Vert x\Vert$). "Length preserved hai" ko appreciate karne ke liye pehle length compute karni aani chahiye.

3. Dot product — ek number jo agreement measure karta hai, likha jaata hai $x^{\mathsf{T}}y$

Yeh tool kyun, koi aur kyun nahi? Hum ek AISA number chahte hain jo humein batae ki do arrows ek hi direction mein, opposite direction mein, ya perpendicular point kar rahe hain. Dot product exactly wahi detector hai:

• agar yeh positive hai, arrows broadly agree karte hain (angle $90^{\circ }$ se kam),
• agar yeh zero hai, woh exactly perpendicular hain (right angle par),
• agar yeh negative hai, arrows broadly disagree karte hain (angle $90^{\circ }$ se zyada).

Picture aur master formula: dot product ke andar ek angle chhupa hai: $x^{\mathsf{T}}y=\Vert x\Vert \Vert y\Vert \cos \theta ,$ jahaan $\theta$ arrows ke beech ka angle hai. Jab arrows perpendicular hote hain, $\theta =90^{\circ }$, $\cos 90^{\circ }=0$, toh poora product $0$ ho jaata hai. Isliye "dot product $=0$" aur "perpendicular" ek hi baat hai.

Topic ko iske zaroorat kyun hai: "columns perpendicular hain" aur "columns ki length 1 hai" — orthonormal ke do halve — dono dot products ke baare mein statements hain.

4. Matrix $Q$ — ek grid jo store karta hai ki arrows kahan jaate hain

Hum $Qx$ likhte hain "machine $Q$ ko arrow $x$ par apply karo" ke liye. Parent ki notation $q_1,q_2,\dots ,q_n$ ka matlab hai "$Q$ ke columns", likha jaata hai $Q=[q_1\mid q_2\mid \cdots \mid q_n]$.

Topic ko iske zaroorat kyun hai: poora subject inhi grids ki ek khaas family ke baare mein hai. Columns mein saari geometry rehti hai.

5. Transpose $Q^{\mathsf{T}}$ — diagonal ke across flip karo

$x^{\mathsf{T}}y$ mein chhota ${}^{\mathsf{T}}$ kyun aata hai: ek plain column vector ek tall $2\times 1$ matrix hai. Ise transpose karne se woh flat hokar ek $1\times 2$ row ban jaata hai. Ek flat row ko ek tall column ke saath rakhkar "row-into-column" multiply karne se exactly ek number milta hai — dot product. Toh $x^{\mathsf{T}}y$ literally hai "$x$ ko flat karo, phir $y$ ke saath combine karo", jo deta hai $x_1{y}_1+x_2{y}_2$. Transpose woh notation hai jo "dot product" ko ordinary matrix multiplication mein badle deta hai.

Topic ko iske zaroorat kyun hai: orthogonal ki definition hi $Q^{\mathsf{T}}Q=I$ hai. Transpose nahi, toh topic nahi.

6. Identity $I$ aur Kronecker delta ${\delta }_{ij}$ — "kuch mat karo" symbols

Topic ko iske zaroorat kyun hai: compact statement "$q_i^{\mathsf{T}}{q}_j={\delta }_{ij}$" ek hi line mein dono orthonormal conditions kehta hai: same column khud ke saath dotted $1$ hai (unit length), alag columns dotted $0$ dete hain (perpendicular).

7. Determinant $\det Q$ — signed area-scaling number

"Signed" kyun? Sign orientation record karta hai. Agar do columns apni original left-to-right ordering maintain karte hain (original axes jaisi), sign $+$ hai. Agar machine ne plane ko flip kar diya — turning ka sense swap kar diya — toh sign $-$ hai. Yahi hai $+1=$ rotation, $-1=$ reflection ki kahani parent note se.

Topic ko iske zaroorat kyun hai: headline result $\det Q=\pm 1$ area ke baare mein ek statement hai. Ek rigid motion area nahi badal sakti, toh scaling factor sirf "area raho, same sign" ($+1$) ya "area raho, flipped" ($-1$) ho sakta hai.

8. Eigenvalue $\lambda$ aur $|\lambda |$ — ek khaas arrow ka khaas stretch

$|\lambda |$ ka matlab $\lambda$ ki size hai sign ignore karke (ya, complex numbers ke liye, zero se uski distance). Ek orthogonal matrix ke liye, koi bhi length kabhi nahi badlti, toh $|\lambda |=1$ hamesha — eigenvalues unit circle par baithe hain. Inhe tum Eigenvalues and eigenvectors mein phir miloge.

Topic ko iske zaroorat kyun hai: parent ki ek key property hai "sabhi eigenvalues $|\lambda |=1$ rakhte hain" — length-preservation ki seedhi echo.

Foundations topic ko kaise feed karte hain

Har ek foundation jo tumne build ki woh aage ki taraf point karti hai: lengths aur dot products orthonormal define karte hain; matrix aur transpose defining condition likhte hain; determinant $\pm 1$ headline deta hai; eigenvalues unit-circle property dete hain. Related gehri raahein Orthonormal bases & Gram–Schmidt, Rotations and reflections in $\mathbb{R}^2$ and $\mathbb{R}^3$, aur Determinants — properties ki taraf jaati hain.

Equipment checklist

Khud ko test karo — right side cover karo aur reveal karne se pehle answer do.