Syllabus

Module 1: Optimization Techniques (12 Hours)
- Preliminary Mathematics: Matrix Calculus, Taylor Series Expansion, Convex Set and Convex Function
- Convex Optimization:
- Unconstrained Optimization by Steepest Descent Method
- Newton’s Method and LMDN Method
- Conjugate Gradient Method
- Constrained Optimization
- Nonconvex Optimization:
- Genetic Algorithm (GA)
- Particle Swarm Optimization (PSO)

-

Module 2: Introduction to Machine Learning (22 Hours)
- Regression:
- Linear and Polynomial Regression
- Classification by Logistic Regression
- Artificial Neural Network:
- Supervised Learning:
- Input-Output Mapping and Classification by Adaptive Linear Model
- Multi-Layer Perceptron: Back-Propagation Learning
- Radial Basis Function Neural Network
- Unsupervised Learning:
- Clustering by Self-Organizing Maps
- Support Vector Machine

-

Module 3: Principal Component Analysis (PCA) (4 Hours)
- Dimension Reduction by PCA

To learn different types of optimization techniques.

To learn structure, training methods and applications of artificial neural networks.

To design fuzzy controllers/fuzzy rule based systems

At the end of the course, students will be able to

Analyze different types of optimization problems and apply appropriate optimization techniques to solve them.

Apply meta-heuristic methods to effectively solve nonconvex optimization problems

Implement single and multi-layer perceptrons for input-output mapping and classification problems.

Apply clustering techniques to categorize datasets based on patterns and relationships.

Utilize Principal Component Analysis (PCA) for dimensionality reduction in large-scale data processing.

Develop and evaluate machine learning models for real-world industrial applications.

Boyd and Vandenberghe, Convex Optimization, Cambridge University Press

S. Haykin, Neural Networks: A Comprehensive Foundation, Pearson

T. J. Ross, Fuzzy Logic with Engineering Application, John Wiley and Sons

V. Kecman, Learning & Soft Computing, Pearson