Einstein summation convention: Visual walkthrough — Hinglish deep dive

4.10.9 · D2 · Maths › Advanced Topics (Elite Level) › Einstein summation convention

Identity ko touch karne se pehle, ek vaada: neeche har symbol pehli baar define hoga, aur ek picture se anchor hoga. Agar tumne zindagi mein kabhi koi index nahi dekha, Step 1 se shuru karo aur padhte raho.

Step 1 — "Index" actually hota kya hai, ek picture pe

KYA kar rahe hain: ek chest of drawers ke teen drawers ko naam de rahe hain. KYUN: poora game is baare mein hai ki kaunse labels repeat hote hain. Repeats ki baat karne ke liye pehle crystal-clear labels chahiye. PICTURE: teen drawers, har ek pe ek number painted hai. Neeche, vector $a$ ek arrow hai jiska har axis pe shadow ek drawer fill karta hai.

Recall "Repeated index = sum" yahan kyun matter karta hai?

Kyunki jis identity ko hum prove kar rahe hain usme left pe ek index $i$ hai jo do baar appear karta hai. ::: Woh double- $i$ ek hidden instruction hai "add over $i = 1, 2, 3$ ", convention ke hisaab se.

Step 2 — Levi-Civita symbol ko ek "spin detector" ki tarah banana

KYA: ek chhoti si machine jo $(1, 2, 3)$ ki ek ordering ko "clockwise / counter-clockwise / degenerate" score karti hai. YEH OBJECT KYUN, AUR PLAIN NUMBER KYUN NAHI? Kyunki cross products aur determinants orientation ki parwah karte hain — tum kis taraf chakkar lagate ho. Ek ratio ya sum orientation yaad nahi rakh sakta; ek sign rakh sakta hai. $ε_{ij k}$ sabse chhoti cheez hai jo "tum kis direction mein spin kar rahe ho" store karta hai. PICTURE: ek circle jisme $1 \to 2 \to 3 \to 1$ arrows hain. Teen labels ko arrows ki direction mein padho → $+ 1$ . Unhe ulti direction mein padho → $- 1$ . Koi bhi doubled label wheel pe nahi baithegi → $0$ .

Step 3 — Product $ε_{ij k} ε_{i l m}$ kya pooch raha hai

KYA: hum do spin-detectors ko multiply kar rahe hain jo apna pehla slot $i$ share karte hain, phir us shared slot ki teen values pe add karte hain. KYUN: $i$ ko share karna aur use sum karna us cheez ka algebraic fingerprint hai jiska matlab hai "in do oriented triples mein kitna overlap hai jab hum pehli direction bhool jaate hain?" Woh overlap ek number hai jo sirf leftovers $j, k, l, m$ pe depend karta hai. PICTURE: do spin-wheels side by side khade hain, ek chalk bridge se jude jo $i$ label hai — yeh bridge summed index hai. Char free stubs $j, k$ aur $l, m$ baahar nikalte hain.

Recall Kitne free indices → answer kaisi cheez hai?

Char free indices $j, k, l, m$ . ::: Toh result ek rank-4 array hai — lekin hum dikhayenge ki yeh purely $δ$ 's ke products se bana hai.

Step 4 — Key fact: $ε$ ek determinant of $δ$ 's hai

KYA: hum sign-machine ko ek determinant ki tarah rewrite karte hain jiske entries sirf Kronecker delta $δ_{ab}$ hain ( $= 1$ agar $a = b$ , warna $0$ ). YEH MOVE KYUN? Ek determinant pehle se sign flip karta hai jab tum do rows swap karo aur vanish hota hai jab do rows equal hon — exactly wahi do rules jo $ε$ ke hain. Toh yeh determinant $ε_{ij k}$ hai, sirf ek lookalike nahi. Yeh ek orientation question ko ordinary matrix algebra mein convert karta hai. PICTURE: $δ$ -entries ka grid. Row $i$ ek unit row hai jiska $1$ column $i$ mein hai; teen aisi rows stacked. Rows $j$ aur $k$ ko swap karna picture ki handedness flip karta hai — aur determinant ka sign.

Step 5 — Do aisi determinants ko multiply karo (product-of-determinants trick)

KYA: humne do $δ$ -grids ko matrix-multiply karke glue kiya, toh har nayi entry hai "first ki row dotted with second ki row" $= δ$ do labels ka. KYUN: $det (M) det (N) = det (M N)$ do orientation-numbers ke product ko ek $3 \times 3$ determinant of overlaps mein badalta hai. Saari geometry ab plain $δ$ 's ke andar hai. PICTURE: do label-strips $(i, j, k)$ aur $(i, l, m)$ ek grid ki tarah cross kiye hue; har cell chalk-blue light up karti hai sirf tab jab uska row-label uske column-label se equal ho.

Step 6 — Shared index $i$ ko sum karo: matrix $3 \to 2$ tak shrink hoti hai

KYA: humne $i$ pe hidden sum chalaya; $i$ se involve hone wala poora pehla row aur column collapse hoga, aur determinant leftovers $j, k, l, m$ mein $3 \times 3$ se $2 \times 2$ drop hoti hai. KYUN: $δ_{ii} = 3$ (dimension) aur substitution rule $δ_{ai} δ_{ib} = δ_{ab}$ sirf wahi tools hain — koi nayi machinery nahi. Result $i$ se free hona chahiye kyunki $i$ sum ho gaya. PICTURE: $3 \times 3$ grid jisme $i$ -row aur $i$ -column greyed out aur fold away hain, ek $2 \times 2$ block bacha hai jiska determinant do diagonal minus do off-diagonal products hai.

Step 7 — Edge & degenerate cases (koi bhi scenario unshown mat chhodna)

YEH KYUN DIKHAO: yeh formula ke har piece ko pin down karte hain — Case A "any repeat → 0" behaviour confirm karta hai, Case B $+$ term confirm karta hai, Case C $-$ term confirm karta hai. Matches ka har combination ab witnessed hai.

Ek picture mein poora summary

Recall Feynman retelling — poora walkthrough plain words mein

Do chhote spinning wheels imagine karo, har ek teen directions se labelled. Ek wheel $+ 1$ output karta hai agar tum uske labels "sahi" tarike se padhte ho, $- 1$ ulte tarike se, aur $0$ agar woh stuck hai (do labels same). Ab dono wheels ko unke pehle spoke pe baandho — woh shared spoke letter $i$ hai jise hum average karne waale hain. Sawaal: us shared direction pe average karne ke baad, dono wheels kitna agree karte hain? Trick yeh hai ki har wheel secretly ek determinant hota hai chhote yes/no switches (deltas, jo kehte hain "yeh do labels match karte hain: yes=1 / no=0"). Do determinants multiply hokar ek bada determinant of match-switches ban jaate hain. Jab hum finally shared spoke $i$ pe add karte hain, woh poora shared row aur column fold up hokar vanish ho jaata hai, ek chhota $2 \times 2$ box rehta hai. Uski value hai "straight matches minus crossed matches": $δ_{j l} δ_{k m} - δ_{j m} δ_{k l}$ . Plus honest agreement hai; minus spin ki bacha hua memory hai. Wahi single line triple cross product ko BAC–CAB rule mein flatten karti hai.

Connections

Levi-Civita symbol aur Kronecker delta — do objects jo yahan glue hue.
Determinants — engine (Steps 4–6): $ε$ $δ$ 's ka ek determinant hai.
Matrix multiplication — Step 5 mein do grids kaise fuse hoti hain.
Cross product aur Vector calculus identities — jahan yeh identity use hoti hai.
Dot product — fused grid ki har entry unit rows ka dot hai.
Parent: Einstein summation convention.