Evolutionary Computation for Single and Multi-Objective Optimization

Evolutionary computation (EC) is a sub-field of computational intelligence that use ideas and get inspiration from natural evolution. It is based on Darwin’s principle of evolution where the population of individuals iteratively performs search and optimization. EC techniques can be applied to optimization, learning, design and many more. This course will concentrate on the concepts, algorithms, hand-calculations, graphical examples, and applications of EC techniques. Topics will be covered include binary and real-coded genetic algorithms, differential evolution, particle swarm optimization, multi-objective optimization and evolutionary algorithms, and statistical assessment. Students will be taught how these approaches identify and exploit biological processes in nature, allowing a wide range of applications to be solved in industry and business. Students will have the opportunity to build and experiment with several different types of EC techniques through-out the course.
INTENDED AUDIENCE :Final and Pre-final year UG students, PG Students and Candidates from IndustriesPREREQUISITES : Elementary Mathematics and ProgrammingINDUSTRIES SUPPORT :All R&D industries that involve design and optimization of product and system

COURSE LAYOUT

Week 1:Introduction and Principles of Evolutionary Computation (EC):Introduction to Optimization, Generalized Formulation, Scope of Optimization via Applications, Characteristic of Optimization Functions;Principles of EC: Natural Evolutional and Genetics, Generalized Framework, Behavior and Typical run of EC, Advantages and LimitationsWeek 2:Binary-Coded Genetic Algorithm (BGA): Introduction, Binary Representation and Decoding, Working Principle of binary coded GA (BGA), BGA on Generalized Framework,Operators, Hand Calculations, Graphical Examples;Week 3:Real-Coded Genetic Algorithm (RGA): Concepts and Need of Real-Coded GA (RGA), Algorithm, RGA on Generalized Framework, Operators, Hand Calculations, Graphical Examples, Case studies;Week 4:Other EC Techniques: Differential Evolution (DE): Introduction, Concepts, Operators, Algorithm, DE on Generalized Framework, Graphical Examples, Case studies; Particle Swarm Optimization (PSO): Introduction, Concepts, Operators, PSO on Generalized Framework, Graphical Examples, Case studies;
Week 5:Constraint Handling Techniques : Generalized Constraint Formulation, Karush Kuhn Tucker (KKT) conditions, Penalty Function Method, Parameter-Less Deb’s Method, Hand Calculations, Graphical Examples, Case studiesWeek 6 Introduction to Multi-Objective Optimization : Introduction, Generalized Formulation, Concept of Dominance and Pareto-optimality, Graphical Examples, Terminologies, Difference with Single-objective optimization, Approaches to multi-objective optimization;Week 7:Classical Multi-Objective Optimization Methods : Classical Multi-Objective Optimization Methods: Weighted- Sum Method, ε-Constraint Method, Weighted Metric Methods, Hand Calculations, Difficulties with Classical approaches, Ideal Multi- Objective Optimization Approach;Week 8:Multi-Objective Evolutionary Algorithms (MOEAs): Introduction, MOEAs on generalized Framework, Algorithms: NSGA-II, SPEA2, Graphical Examples, Case Studies; Hypervolume Indicator (HV) for Performance Assessment