1 / 19

MKI44: Evolutionary Algorithms

MKI44: Evolutionary Algorithms. Organisation. Teachers. Ida Sprinkhuizen-Kuyper Room B.02.39 E-mail: i.kuyper@donders.ru.nl Phone: 024-3616126 URL: http://www.nici.ru.nl/~idak

candie
Télécharger la présentation

MKI44: Evolutionary Algorithms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. MKI44: Evolutionary Algorithms Organisation MKI44: EAs

  2. Teachers • Ida Sprinkhuizen-KuyperRoom B.02.39E-mail: i.kuyper@donders.ru.nlPhone: 024-3616126URL: http://www.nici.ru.nl/~idak • Pim HaselagerE-mail: w.haselager@donders.ru.nlRoom B.02.40 Phone: 024-3616066URL: http://www.nici.ru.nl/~haselag • Ruud BarthE-mail: rudoros@gmail.com MKI44: EAs

  3. Evolutionary Algorithms • Evolutionary Algorithms • Set up: • Theory • Self study • Presenting summaries • Discussion • Practice • Project using/studying EAs MKI44: EAs

  4. Goals • This course contributes to the following final qualifications of a master AI: • 1: Knowledge and understanding of AI • 4: Knowledge and understanding of different model types • 5: Analysing problems • 6: Research skills • 11: Learning skills MKI44: EAs

  5. Material • Book: A.E. Eiben and J.E. Smith, Introduction to Evolutionary Computing, corrected 2nd printing, 2007  • Websites:book + material:http://www.cs.vu.nl/~gusz/ecbook/ecbook.htmlcourse:http://www.ai.ru.nl/aicourses/mki44ECJ:http://www.cs.gmu.edu/~eclab/projects/ecj/ MKI44: EAs

  6. What are Eas? • Stochastic, population-based, general applicable problem-solving algorithms, inspired by natural evolution • Survival of the fittest MKI44: EAs

  7. General scheme MKI44: EAs

  8. Typical EA MKI44: EAs

  9. Problem type 1 : Optimisation • We have a model of our system and seek inputs that give us a specified goal • e.g. • time tables for university, call center, or hospital • design specifications, etc etc MKI44: EAs

  10. Problem type 2: Modelling • We have corresponding sets of inputs & outputs and seek model that delivers correct output for every known input • Evolutionary machine learning MKI44: EAs

  11. Problem type 3: Simulation • We have a given model and wish to know the outputs that arise under different input conditions • Often used to answer “what-if” questions in evolving dynamic environments • e.g. Evolutionary economics, Artificial Life MKI44: EAs

  12. Global schedule • Today (8-9): Introduction • Next week: • Tuesday (15-9): More about evolution (Pim) • Wednesday (16-9): Working with ECJ (Ruud) • Upto 3-11: Studying the book, designing a project • Tuesdays 22-9 till 3-11: short presentations of the chapters and discussion • Wednesdays: Practical work with ECJ • Upto 19-1-2010: Project, guest lectures • 19-1-2010: Presentation/demonstration of the projects MKI44: EAs

  13. Chapters • 22-9 • 29-9 • 29-9 • 6-10 • 6-10 • 13-10 • 13-10 • 20-10 • 20-10 • 27-10 • 27-10 • 3-11 • 3-11: Pim/Ida • Introduction • What is an Evolutionary Algorithm? • Genetic Algorithms • Evolution Strategies • Evolutionary Programming • Genetic Programming • Learning Classifier Systems • Parameter Control in Evolutionary Algorithms • Multi-Modal Problems and Spatial Distribution • Hybridisation with Other Techniques: Memetic Algorithms • Theory • Constraint Handling • Special Forms of Evolution • Working with Evolutionary Algorithms • Summary MKI44: EAs

  14. Organisation • We will randomly distribute the chapters • For presenting a concise summary: 1 or 2 students • For formulating some discussion questions: 2 or 3 students • All students have to study the chapters before the lecture and should be involved in questions and discussions during the lectures • Goal of studying the book is to learn the possibilities of the different forms of Eas, learning how to use the terminology correctly, how to choose important parameters, etc. MKI44: EAs

  15. The project • Groups of 2 or 3 students • Project proposal: deadline 3-11 • Research question • Motivation for EAs • Experimental set up: • Representation • Fitness function • Type(s) of Eas • … MKI44: EAs

  16. Examination • The result of the course is determined by the project • The Project will be judged on • The presentation/demonstration (20%) • Project proposal, design, implementation, originality (40%) • The report (40%) • Motivation of choices (representation, fitness function, type of Eas, …) • Correct use of EA terminology • Statistical analysis of the results MKI44: EAs

  17. Ideas for projects • Aspects of a project: • Task • EA • Task types • Optimizing (scheduling, robot controller, …) • Modeling (datamining, bci, …) • Simulation (mirror neurons, artificial societies, …) MKI44: EAs

  18. Ideas for projects (2) • EAs • Genetic algorithms • Genetic programming • Constraint satisfaction • Coevolution • … MKI44: EAs

  19. Examples • WEIRD webpagehttp://www.ru.nl/ai/onderwijs/stages_scripties/weird/ • Student projects: Many examples of projects • Demo of an EA for evolving a robot controller for a box-pushing task MKI44: EAs

More Related