RTMNU · Final Year B.Tech Electronics & Telecommunication Engineering · CBCS
Fundamentals of Machine Learning
BEETC703PE · Elective V · Semester VII · 3 credits · College 30 + university 70
One of the Elective V options.
Unit 1: Introduction and Supervised Learning
8 hoursExamples of machine learning — associations, classification, regression, unsupervised and reinforcement learning; learning a class from examples, the VC dimension, PAC learning, noise, multiple classes, regression, model selection and generalisation.
Examined in the 70-mark end-semester paper
- Introduction to machine learning
- Types of machine learning
- Hypothesis space and inductive bias
- VC dimension
- PAC learning and sample complexity
Unit 2: Bayesian Decision Theory and Parametric Methods
7 hoursClassification, losses and risks, discriminant functions and association rules; maximum likelihood estimation for Bernoulli, multinomial and Gaussian densities; bias and variance of an estimator, the Bayes estimator, parametric classification and regression, and model selection.
Examined in the 70-mark end-semester paper
- Bayesian decision theory: losses, risks and discriminant functions
- Method of moments and maximum likelihood
- Regression basics and the bias–variance trade-off
- Cross-validation and micro and macro averaging
Unit 3: Multivariate Methods and Dimensionality Reduction
7 hoursMultivariate data, parameter estimation, missing values, the multivariate normal distribution and multivariate classification and regression; subset selection, PCA, feature embedding, factor analysis, SVD and matrix factorisation, multidimensional scaling, LDA and canonical correlation analysis.
Examined in the 70-mark end-semester paper
- Feature selection: sequential forward and backward
- PCA and LDA
- Factor analysis, Isomap and locally linear embedding
- Singular value decomposition
- Multidimensional scaling and matrix factorisation
Unit 4: Clustering and Nonparametric Methods
8 hoursMixture densities, k-means, the EM algorithm, latent-variable mixtures, spectral and hierarchical clustering and choosing the number of clusters; nonparametric density estimation with histogram, kernel and k-NN estimators, nonparametric classification, condensed nearest neighbour and outlier detection.
Examined in the 70-mark end-semester paper
- K-means and K-medoids
- Expectation maximisation
- Spectral clustering
- Clustering and hierarchical clustering
- Nonparametric density estimation: histogram and kernel estimators
- K-nearest neighbours
- Outlier detection methods
Unit 5: Decision Trees and Linear Discrimination
6 hoursUnivariate, classification and regression trees, pruning and rule extraction, multivariate trees; generalising the linear model, the geometry of linear discriminants, pairwise separation, gradient descent and logistic discrimination for two and more classes.
Examined in the 70-mark end-semester paper
- Decision tree induction
- Gradient descent
- Linear and logistic regression
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).