datarekha

Formula & Trap Revision Sheets

One compact, scannable formula-and-trap sheet per subject — the night-before-the-exam reference where the Traps lines flag exactly where marks leak.

12 min read Intermediate GATE DA Lesson 122 of 122

Probability & Statistics

Formulas to know cold

IdeaFormula
UnionP(A∪B) = P(A) + P(B) − P(A∩B)
Bayes`P(H
VarianceVar(X) = E[X²] − (E[X])²
CovarianceCov(X,Y) = E[XY] − E[X]·E[Y]
Bernoulli(p)mean p, variance p(1−p)
Binomial(n,p)mean np, variance np(1−p)
Uniform(a,b)mean (a+b)/2, variance (b−a)²/12
Exponential(λ)mean 1/λ, variance 1/λ²
Poisson(λ)mean λ, variance λ (equal)
Normal(μ,σ²)mean μ, variance σ²
Standardizez = (x − μ)/σ (divide by σ, the SD)
CLTsample-mean SD = σ/√n (the standard error)

Traps

Linear Algebra

Formulas to know cold

IdeaFact
Determinantdet(A) = Πλᵢ (product of eigenvalues)
Tracetrace(A) = Σλᵢ (sum of eigenvalues)
Rank–nullityrank(A) + nullity(A) = number of COLUMNS
Invertibledet ≠ 0 ⇔ full rank ⇔ 0 is NOT an eigenvalue
Symmetricreal eigenvalues, orthogonal eigenvectors
IdempotentP² = P ⇒ eigenvalues are 0 or 1 only
Orthogonal QQᵀQ = I, det(Q) = ±1, columns orthonormal
Singular valuesσᵢ = √(eigenvalues of AᵀA), always ≥ 0
Quadratic form maxmax xᵀAx on ‖x‖ = 1 is the largest eigenvalue (min is the smallest)

Traps

Calculus & Optimization

Formulas to know cold

IdeaFact
Standard limitssin x / x → 1 and (eˣ − 1)/x → 1 as x → 0
L’Hôpitalfor 0/0 or ∞/∞ only, differentiate top and bottom
2nd-derivative testf'' > 0 ⇒ local min; f'' < 0 ⇒ local max; f'' = 0 inconclusive
Closed-interval optimisationcompare critical points AND both endpoints
Convexityconvex ⇒ any local min is the global min
Taylor coefficientf⁽ⁿ⁾(0) = n! · (coefficient of xⁿ) in the series
Maclaurineˣ = Σ xⁿ/n!; sin x = x − x³/3! + …; cos x = 1 − x²/2! + …

Traps

Programming & DSA

Formulas to know cold

AlgorithmTimeNotes
Bubble / Insertion / SelectionΘ(n²)bubble & insertion stable; selection NOT stable
MergesortΘ(n log n) alwaysstable; needs O(n) extra space
Quicksortavg Θ(n log n), worst Θ(n²)in place, NOT stable
Binary searchΘ(log n)requires a SORTED array
Hashing (open addressing)expected probes ≈ 1/(1 − α)α = n/m load factor
BFS / DFSO(V+E)BFS uses a queue, DFS uses a stack
Tree reconstructioninorder + (pre or post) rebuilds a binary tree uniquely

Traps

Databases & Warehousing

Formulas to know cold

IdeaFact
Candidate keya minimal attribute set whose closure is all attributes
Attribute closurerepeatedly add RHS of any FD whose LHS is already in the set
2NFno non-prime attribute depends on part of a candidate key
3NFfor every FD X → A: X is a superkey OR A is prime
BCNFfor every non-trivial FD X → A: X must be a superkey (no exception)
Lossless-jointhe common attribute set R1 ∩ R2 is a superkey of one piece
Index choicehash index for equality; B+-tree for range / ordered scans
Scalingmin-max maps to [0,1]; z-score gives mean 0, SD 1
Schemastar = denormalised dimensions; snowflake = normalised (more joins)
Measuresdistributive (SUM, COUNT, MIN, MAX); algebraic (AVG, stddev); holistic (MEDIAN, MODE)

Traps

Machine Learning

Formulas to know cold

IdeaFormula
Least squaresminimize Σ(yᵢ − ŷᵢ)²; normal equation w = (XᵀX)⁻¹Xᵀy
RidgeΣ(yᵢ − ŷᵢ)² + λ·‖w‖² (L2 penalty); ↑λ ⇒ ↑bias, ↓variance
Precision / Recallprecision = TP/(TP+FP), recall = TP/(TP+FN)
F1F1 = 2·precision·recall / (precision + recall)
Sigmoidσ(z) = 1/(1 + e^−z) maps to (0,1)
Naive Bayesposterior ∝ prior · Π likelihoods
k-meansalternate assign-to-nearest-centroid, then update centroid to the mean
MLP paramsper layer a·b weights + b biases; total summed over layers
ReLUmax(0, x) — continuous, NOT differentiable at 0
PCAcomponents orthogonal; variance explained = eigenvalue / sum of eigenvalues
LOOCVnumber of folds = n (leave one out each time)

Traps

Artificial Intelligence

Formulas to know cold

IdeaFact
Uninformed searchBFS complete + optimal (unit costs); DFS neither; UCS optimal on costs; IDDFS combines BFS-optimality with DFS-space
Admissible heuristich(n) ≤ true cost, i.e. never overestimates
A*expand by f = g + h; admissible h ⇒ optimal
Minimaxback values up: MAX takes the max of children, MIN the min
Alpha-betaprunes branches but returns the SAME value as full minimax
EntailmentX ⊨ Y if and only if X ∧ ¬Y is unsatisfiable
FOL translation”all A are B” is ∀x A(x) ⇒ B(x); “some A is B” is ∃x A(x) ∧ B(x)
Bayes netjoint `= Π P(node
Inferencevariable elimination is EXACT; sampling is APPROXIMATE

Traps

General Aptitude

Formulas to know cold

IdeaFact
Successive percentagesMULTIPLY the factors: +20% then −10% is ×1.20×0.90, not +10%
Percentage changealways measured over the BASE (the original value)
Ratiosscale every part by the same multiplier; compare like-for-like
The four GA bucketsquantitative, verbal, analytical reasoning, spatial reasoning

Traps


Next, prove it under the clock: run the timed mocks in the Exam Lab, then use /gate-da/review to re-test whatever these sheets reminded you was shaky.

Sign in to track your progress

Completed lessons, your XP, level, and streak save to your account — it's free and takes a few seconds.

Related lessons

Skip to content