GATE Data Science 2025 - Formula & Revision
Checklist
1. Probability & Statistics
• Bayes Theorem: P(A|B) = [P(B|A) * P(A)] / P(B)
• Expectation: E[X] = Σx * P(x)
• Variance: Var(X) = E[X²] - (E[X])²
• Standard Distributions: Binomial, Poisson, Normal, Exponential
• Central Limit Theorem basics
2. Linear Algebra
• Matrix multiplication rules
• Determinant properties: det(AB) = det(A)det(B)
• Rank of a matrix = dimension of column space
• Eigenvalue equation: Av = λv
• Orthogonality: u·v = 0
3. Optimization
• Gradient Descent update: θ = θ - α ∇J(θ)
• Convex function definition
• First-order condition: ∇f(x*) = 0
4. Machine Learning
• Linear Regression: y = Xβ + ε
• Logistic Regression: P(y=1) = 1 / (1 + e^(-z))
• Naive Bayes: P(y|x) ∝ P(y) ∏ P(xi|y)
• SVM: Maximize margin = 2/||w||
• Decision Tree: Information Gain = Entropy(parent) - weighted avg(Entropy(children))
• K-means: minimize within-cluster variance
5. Data Structures & Algorithms
• Time complexity (Big-O) basics: O(1), O(log n), O(n), O(n log n), O(n²)
• Sorting complexities: QuickSort O(n log n), MergeSort O(n log n), HeapSort O(n log n)
• Graph traversal: BFS (Queue), DFS (Stack/Recursion)
• Binary Search Tree: Avg O(log n), Worst O(n)
• Hashing: O(1) average, O(n) worst
6. Databases & SQL
• ACID properties (Atomicity, Consistency, Isolation, Durability)
• Normalization: 1NF, 2NF, 3NF basics
• Joins: INNER, LEFT, RIGHT, FULL
• Indexing improves SELECT speed, slows INSERT/UPDATE
• Transaction: COMMIT, ROLLBACK
7. Operating Systems (Basics)
• Process vs Thread
• CPU Scheduling: FCFS, SJF, Round Robin
• Deadlock conditions (Mutual exclusion, Hold & Wait, No preemption, Circular wait)
• Paging vs Segmentation
• Virtual Memory basics
8. General Aptitude
• Average = (Sum of values) / (Number of values)
• Simple Interest = (P × R × T)/100
• Compound Interest = P(1+R/100)^T - P
• Speed = Distance/Time
• Ratio, Proportion, Percentages, Profit & Loss
