Course detail

Applied Evolutionary Algorithms

FIT-EVOAcad. year: 2018/2019

Overview of principles of stochastic search techniques: Monte Carlo (MC) methods, evolutionary algorithms (EAs). Detailed explanation of selected MC algorithms: Metropolis algorithm, simulated annealing, their application for optimization and simulation. Overview of basic principles of EAs: evolutionary programming (EP), evolution strategies (ES), genetic algorithms (GA), genetic programming (GP).  Advanced EAs and their applications: numerical optimization, differential evolution (DE), social algoritmhs: ant colony optimization (ACO) and particle swarm optimization (PSO). Multiobjective optimization algorithms. Applications in solving engineering problems and artificial intelligence.

Learning outcomes of the course unit

Ability of problem formulation for the solution on the base of evolutionary computation. Knowledge of analysis and design methods for evolutionary algorithms.

Prerequisites

Not applicable.

Co-requisites

Not applicable.

Recommended optional programme components

Not applicable.

Recommended or required reading

  • Luke, S.: Essentials of Metaheuristics. Lulu, 2015, ISBN 978-1-300-54962-8
  • Jansen, T.: Analyzing Evolutionary Algorithms. Springer-Verlag, Berlin Heidelberg, 2013, ISBN 978-3-642-17338-7
  • Kvasnička, V., Pospíchal, J., Tiňo, P.: Evolučné algoritmy. Vydavatelství STU Bratislava, Bratislava, 2000, ISBN 80-227-1377-5
  • Oplatková, Z., Ošmera, P., Šeda, M., Včelař, F., Zelinka, I.: Evoluční výpočetní techniky - principy a aplikace. BEN - technická literatura, Praha, 2008, ISBN 80-7300-218-3

  • Brabazon, A., O'Neill, M., McGarraghy, S.: Natural Computing Algorithms. Springer-Verlag Berlin Heidelberg, 2015, ISBN 978-3-662-43630-1
  • Eiben, A.E., Smith, J.E.: Introduction to Evolutionary Computing, 2nd ed. Springer-Verlag Berlin Heidelberg, 2015, ISBN 978-3-662-44873-1
  • Jansen, T.: Analyzing Evolutionary Algorithms. Springer-Verlag, Berlin Heidelberg, 2013, ISBN 978-3-642-17338-7
  • Talbi, E.-G.: Metaheuristics: From Design to Implementation. Wiley, Hoboken, New Jersey, 2009, ISBN 978-0-470-27858-1
  • Bäck, T.: Evolutionary Algorithms in Theory and Practice. Oxford University Press, Oxford, 1996, ISBN 978-0195099713

Planned learning activities and teaching methods

Not applicable.

Assesment methods and criteria linked to learning outcomes

Evaluated practices, project. In the case of a reported barrier preventing the student to perform scheduled activity, the guarantor can allow the student to perform this activity on an alternative date.
Exam prerequisites:
None.

Language of instruction

Czech

Work placements

Not applicable.

Course curriculum

    Syllabus of lectures:
    1. Introduction, terminology, principles of stochastic search algorithms.
    2. Monte Carlo method and variants (Metropolis algorithm, Simulated Annealing).
    3. Basic evolutionary algorithms (Evolutionary Programming, Evolution Strategies).
    4. Genetic algorithms (control parameters, genetic operators).
    5. Genetic programming and symbolic regression.
    6. Case studies (design of algorithms and electronic circuits).
    7. Differential evolution (numerical optimization, engineering case study).
    8. Advanced evolutionary techniques (Estimation of Distribution Algorithms).
    9. Multiobjective evolutionary algorithms (basic techniques, engineering case study).
    10. Advanced multiobjective evolutionary algorithms.
    11. Parallel evolutionary algorithms and coevolutionary algorithms.
    12. Evolutionary development and grammatical evolution.
    13. Social computing algorithms (Particle Swarm Optimization, Ant algorithms).

    Syllabus of laboratory exercises:
    • Basic concepts of evolutionary computing, typical problems, solution of a technical task using a variant of Metropolis algorithm.
    • Evolutionary algorithms in engineering areas, optimization of electronic circuits using genetic algorithm.
    • Evolutionary design using genetic programming.
    • Differential evolution, estimation of distribution algorithms.
    • Optimization using social computing algorithms.
    • Solution of selected problems of statistical mechanics.

    Syllabus - others, projects and individual work of students:
    Solution of individual selected topic either by:
    • implementing a given application from the field of evolutionary computation or
    • study of a given paper, presentation of main ideas.
    By agreement there is a possibility to include solution of the project from other course (e.g. BIN) to EVO if its topic belongs to evolutionary computation.

Aims

Survey about actual optimization techniques and evolutionary algorithms for solution of complex, NP complete problems. To learn how to solve typical complex tasks from engineering practice using evolutionary techniques.

Classification of course in study plans

  • Programme IT-MGR-2 Master's

    branch MBI , any year of study, summer semester, 5 credits, compulsory-optional
    branch MPV , any year of study, summer semester, 5 credits, compulsory-optional
    branch MGM , any year of study, summer semester, 5 credits, elective
    branch MSK , any year of study, summer semester, 5 credits, elective
    branch MIS , any year of study, summer semester, 5 credits, elective
    branch MBS , any year of study, summer semester, 5 credits, elective
    branch MIN , any year of study, summer semester, 5 credits, elective
    branch MMI , any year of study, summer semester, 5 credits, elective
    branch MMM , any year of study, summer semester, 5 credits, elective

Type of course unit

 

Lecture

26 hours, optionally

Teacher / Lecturer

Syllabus

  1. Principles of stochastic search algorithms.
  2. Monte Carlo methods.
  3. Evolutionary programming and evolution strategies.
  4. Genetic algorithms.
  5. Genetic programming.
  6. Models of computational development.
  7. Statistical evaluation of experiments.
  8. Ant colony optimization.
  9. Particle swarm optimization.
  10. Differential evolution.
  11. Applications of evolutionary algorithms.
  12. Fundamentals of multiobjective optimization.
  13. Advanced algorithms for multiobjective optimization.

Exercise in computer lab

12 hours, compulsory

Teacher / Lecturer

Project

14 hours, compulsory

Teacher / Lecturer

Syllabus

Realisation of individual topics from the area of evolutionary computation.

eLearning