Previous Year Question Papers for AIML & DSE (All Semesters)

Linear Algebra, Orthogonal Projections, Gram-Schmidt, Eigen Decomposition, SVD, Multivariate Calculus, Hessian Matrices, Convex Optimization, Lagrange Multipliers, and KKT Conditions.

Probability Distributions, Bayes Theorem, Sampling Distributions, Point & Interval Estimation, Hypothesis Testing (Z-test, t-test, ANOVA, Chi-Square), and Linear Regression.

Supervised & Unsupervised Algorithms, Decision Trees (ID3, C4.5, Gini), Support Vector Machines (Lagrangian Duality, Kernels), Ensembles (Bagging, Boosting, Random Forests), and Clustering.

Multi-Layer Perceptrons, Backpropagation Derivations, Activation Functions, CNNs (Convolution arithmetic, ResNet, VGG), RNN/LSTM/GRU Architectures, Regularization (Dropout, BatchNorm), and Optimizers (Adam, RMSprop).

Classical Search (A*, IDA*, heuristic admissibility & consistency), Adversarial Game Trees (Minimax, Alpha-Beta pruning), CSPs, First-Order Logic Inference (Resolution, Unification, Forward/Backward Chaining), Bayesian Networks, Fuzzy Logic, and Genetic Algorithms.

Markov Decision Processes (MDP), Bellman Expectation & Optimality Equations, Tabular Policy & Value Iteration, Monte Carlo Methods, TD(0), SARSA, Q-Learning, Function Approximation (DQN, Double DQN, Dueling DQN), Policy Gradients (REINFORCE), and Actor-Critic (A2C, PPO).

Text Preprocessing, N-gram Language Modeling (Good-Turing, Kneser-Ney), Part-of-Speech Tagging (HMMs, Viterbi Algorithm), Word Embeddings (Word2Vec CBOW & Skip-Gram, GloVe), Seq2Seq, Attention Mechanisms, Transformer Architecture, BERT, and LLMs.

Inverted Index Construction, Boolean Query Processing, Vector Space Model (TF-IDF, Cosine Similarity), BM25 Ranking, Evaluation Metrics (Precision@k, Recall, MAP, MRR, NDCG), Index Compression (Gamma, Variable-Byte), and Link Analysis (PageRank, HITS).

Feature Stores, Feature Engineering Pipelines, Streaming Data Ingestion (Kafka, Flink), Distributed Storage (HDFS, S3, Parquet), Data Versioning (DVC, Delta Lake), Relational & NoSQL Data Modeling, and Data Quality Assurance for Production ML.

Bayesian Networks (D-separation, I-maps), Markov Random Fields, Exact Inference (Variable Elimination, Clique Trees, Message Passing), Approximate Inference (Gibbs Sampling, MCMC), and Parameter Estimation with Expectation-Maximization.