Lecture #9: Fuzzy Logic & Neural Nets

Lecture #9: Fuzzy Logic & Neural Nets

<p> Monday February 2, 2004</p><p>DSES-4810-01 Intro to COMPUTATIONAL INTELLIGENCE & SOFT COMPUTING</p><p>Instructor: Prof. Mark J. Embrechts (x 4009 or 371-4562) ([email protected]) Office Hours: Thursday 10-11 am (CII5217) Or by appointment. Class Time: Monday/Thursday: 8:30-9:50 (Amos eaton hall 216) TEXT (optional): J. S. Jang, C. T. Sun, E. Mizutani, “Neuro-Fuzzy and Soft Computing,” Prentice Hall, 1996. (1998) ISBN 0-13-261066-3</p><p>LECTURES 6&7: INTRO to GENETIC ALGORITHMS & EVOLUTIONARY COMPUTING</p><p>Genetic algorithms and evolutionary computing are general-purpose optimization algorithms that are inspired from biological evolution. Evolutionary computing is the third tier of soft computing (the others being neural networks and fuzzy logic). First proposed in the sixties by Prof. John Holland (U. Michigan), they bloomed in the late eighties and the first IEEE journal on evolutionary computing appeared in 1998. Evolutionary computing is a broader framework than genetic algorithms.</p><p>Genetic algorithms are a derivative free stochastic optimization method (cfr. simulated annealing) based loosely on the concepts of natural selection and evolutionary processes. Genetic Algorithms are nowadays often applied in Operations Research applications for rapidly estimating approximate results.</p><p>Genetic algorithms find global optima without relying on gradients, are robust and often provide surprisingly good results fast. GAs and evolutionary computation have several important industrial applications (e.g., job scheduling, traveling salesman type of problems) because of the following desirable characteristics: Only a cost function is required, no derivatives Ideal parallel search procedure Stochastic in nature, less likely to be trapped in local minima For designing a GA the following steps are necessary: Encoding scheme Fitness Function Selection Cross-Over Mutation Software: An excellent public domain package for exprt C++ users is GALib A good freeware MATLAB package is GAOT A nice, easy to use and simple C code can be ftp’d via: ftp.uncc.edu, directory coe/evol, file prog.c</p><p>1 Handout:</p><p>1. Philip D. Wasserman, Advanced Methods in Neural Computing, Chapter 5: Genetic Algorithms, Van Nostrand Reinhold (1993).</p><p>H omework #2: </p><p>Find an article related to your project and prepare a 1-2 page abstract summarizing the article. Prepare 10-15 slides (on paper) as if you were going to present the paper. Attach a copy of the paper and, if relevant, important supporting articles as well. Make sure you reference the paper in the abstract. Note: the choice of a relevant paper subject and the quality of the selected paper will influence your grade.</p><p>Deadlines</p><p>January 22 Homework Problem #1 (web browsing) January 29 Project Proposal February 16 Homework Problem #1 (paper)</p><p>2</p>

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    2 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us