Machine Learning Techniques (B.Tech)

Departmental Elective-II: supervised and unsupervised learning, decision trees, neural networks and SVM.

6 MonthsDuration 5Chapters30Topics

Enquire now

Syllabus · पाठ्यक्रम

Download PDF

01Unit 1: Introduction to Machine Learning 6 topics

Learning and types of learning
Well-defined learning problems and designing a learning system
History of machine learning
Overview of ML approaches: ANN, clustering, RL, decision trees, Bayesian networks, SVM, GA
Issues in machine learning; data science vs machine learning
Overfitting, underfitting and bias-variance

02Unit 2: Regression, Bayesian Learning and SVM 6 topics

Linear regression
Logistic regression
Bayes theorem, concept learning and Bayes optimal classifier
Naive Bayes and Bayesian belief networks
EM algorithm
SVM: hyperplane, margin, kernels (linear, polynomial, Gaussian)

03Unit 3: Decision Trees and Instance-based Learning 6 topics

Decision tree learning and inductive bias
Entropy and information gain, ID3
Issues in decision trees: overfitting, pruning
k-nearest neighbour learning
Locally weighted regression and radial basis function networks
Case-based learning

04Unit 4: Artificial Neural Networks and Deep Learning 6 topics

Perceptron and multilayer perceptron
Gradient descent and delta rule
Back-propagation
Self-organising maps
Deep learning and CNN: convolution, activation, pooling, fully connected layers
Applications of CNN

05Unit 5: Reinforcement Learning and Genetic Algorithms 6 topics

Reinforcement learning task and examples
Markov decision process
Q-learning function and algorithm
Genetic algorithm components and cycle
Selection, crossover and mutation
Genetic programming and applications

About

You will be able to

Define well-posed learning problems and compare supervised, unsupervised and reinforcement learning
Apply linear and logistic regression, naive Bayes and support vector machines
Build decision trees with ID3 using entropy and information gain and apply k-nearest neighbour
Explain perceptrons, back-propagation and the layers of a convolutional neural network
Explain Markov decision processes and Q-learning
Explain genetic algorithms: selection, crossover and mutation

Who it is for

B.Tech Computer Science & Engineering students of Semester 5 (AKTU)
Students of other universities studying the same subject who want unit-wise coaching

What you need

Engineering mathematics (linear algebra, probability)
Python programming basics

Questions

Which syllabus does the Machine Learning Techniques course follow?

It follows the AKTU (Dr. A.P.J. Abdul Kalam Technical University, Lucknow) B.Tech syllabus for Machine Learning Techniques. If your own university's syllabus differs, the centre follows your university syllabus.

Will this help in the university examination?

Yes. Teaching is unit-wise as per the AKTU scheme, with previous-year question practice, MCQ tests and doubt sessions before the semester exam.

Is coding taught in this theory elective?

Concepts and numericals are taught for the exam, and each algorithm is also demonstrated in Python with scikit-learn.