Una-May O'Reilly: Difference between revisions
From Santa Fe Institute Events Wiki
| (9 intermediate revisions by the same user not shown) | |||
| Line 7: | Line 7: | ||
[http://www.santafe.edu/events/workshops/index.php/CSSS_2007_Santa_Fe Back to CSSS Wiki Main Page] | [http://www.santafe.edu/events/workshops/index.php/CSSS_2007_Santa_Fe Back to CSSS Wiki Main Page] | ||
== What Should I Lecture On?== | == What Should I Lecture On?== | ||
| Line 17: | Line 19: | ||
* Are there any topics or concepts in evolutionary algorithms that are burning a hole in your head that you'd like me to cover? | * Are there any topics or concepts in evolutionary algorithms that are burning a hole in your head that you'd like me to cover? | ||
== | == Answers to My [[Survey]] on CSSS-07 Participants' Knowledge of EvComp == | ||
==Background Material is posted with Lecture material [http://www.santafe.edu/events/workshops/index.php/CSSS_2007_Santa_Fe-Readings here] == | |||
==Evolutionary Algorithm Software== | |||
* Ken De Jong at George Mason University is the source of [http://www.cs.gmu.edu/~eclab/projects/ec_courseware/ the software I demonstrate in the first lecture]. | |||
The two demonstrations in the lecture were defined by two .cfg files that should be executed in the EC2 directory. I've described them below. Cut and pasted the text into a file of the appropriate name: | |||
* [[File:evss.cfg.txt|evss.cfg]] | |||
* [[File:evgen.cfg.txt|evgen.cfg]] | |||
Homework? Never! But for some people the best way to understand a genetic algorithm or any other sort of Evolutionary Algorithm is to implement or use one. I insist that my MIT students write at least one from scratch but, as a project grows, it is often more efficient not to frequently reinvent the wheel. Google or any other internet search engine suffices to find software implementations. They come in an assortment of languages, vary in age, support and popularity. Here's a sample of ones I tend to hear of being used in the community: | Homework? Never! But for some people the best way to understand a genetic algorithm or any other sort of Evolutionary Algorithm is to implement or use one. I insist that my MIT students write at least one from scratch but, as a project grows, it is often more efficient not to frequently reinvent the wheel. Google or any other internet search engine suffices to find software implementations. They come in an assortment of languages, vary in age, support and popularity. Here's a sample of ones I tend to hear of being used in the community: | ||
| Line 42: | Line 46: | ||
[[Image:Unamay-self-sm-may07.jpg|thumb|caption]] | [[Image:Unamay-self-sm-may07.jpg|thumb|caption]] | ||
== Experimental Space == | == Experimental Space == | ||
Latest revision as of 11:31, 21 June 2007
I am from the Computer Science and Artificial Intelligence Lab, (CSAIL), MIT where I lead EVO-DesignOpt.
My bio and other stuff on my web site tell you about me, my research agenda and various projects. A little extra related info: I was an SFI Doctoral fellow from 1993-1995. (I've been at MIT ever since.) And, I met my husband at SFI. In fact, he was the first SFI'r I met when I arrived. How about that for good fortune!
What Should I Lecture On?
I will be posting a short set of questions to the CSSS email list.
- I want to know your experience level with respect to genetic algorithms, genetic programming, evolutionary strategies, evolutionary programming, particle swarm optimization, BOA, hBOA, grammatical evolution or any other Evolutionary Algorithm. Responses might be one of "expert, knowledgeble, ignorant" and one of "develop, use, don't use".
- I want to know, if you know, how you have used or anticipate using evolutionary computation in your field of research.
- Are there any topics or concepts in evolutionary algorithms that are burning a hole in your head that you'd like me to cover?
Answers to My Survey on CSSS-07 Participants' Knowledge of EvComp
Background Material is posted with Lecture material here
Evolutionary Algorithm Software
- Ken De Jong at George Mason University is the source of the software I demonstrate in the first lecture.
The two demonstrations in the lecture were defined by two .cfg files that should be executed in the EC2 directory. I've described them below. Cut and pasted the text into a file of the appropriate name:
Homework? Never! But for some people the best way to understand a genetic algorithm or any other sort of Evolutionary Algorithm is to implement or use one. I insist that my MIT students write at least one from scratch but, as a project grows, it is often more efficient not to frequently reinvent the wheel. Google or any other internet search engine suffices to find software implementations. They come in an assortment of languages, vary in age, support and popularity. Here's a sample of ones I tend to hear of being used in the community:
- SFI is looking into buying the commerical Matlab GA and Direct Search toolbox. It is accompanied by a short article that compares GAs on the optimization of a few functions.
- A GP S/W Directory
- ECJ: A Java-based Evolutionary Computation and Genetic Programming Research Toolbox
- Open BEAGLE: an Evolutionary Computation (EC) framework in C++.
- PyLEA: Python Lover's Evolutionary Algorithms
- PerlGP
- GP in Matlab
Unamay 05:02, 7 June 2007 (MDT)
Experimental Space
Here's a link to the Main page.
Section
Subsection
Subsubsection
- Unordered lists are easy to do:
- Start every line with a star.
- More stars indicate a deeper level.
- Previous item continues.
- A new line
- Start every line with a star.
- in a list
marks the end of the list.
- Of course you can start again.
- Numbered lists are:
- Very organized
- Easy to follow
A new line marks the end of the list.
- New numbering starts with 1.
- A colon (:) indents a line or paragraph.
A newline starts a new paragraph.
Often used for discussion on talk pages.
- We use 1 colon to indent once.
- We use 2 colons to indent twice.
- 3 colons to indent 3 times, and so on.
- We use 2 colons to indent twice.