Revision 7412d2d977fa676cff6583ccb2ad031090606c1d (click the page title to view the current version)
Optimisation and GA
Reading
The textbook chapters are available at BlackBoard.
- Haupt and Haput Chapter 1 (cursory)
- we will not dwell on the various optimisation algorithms discussed here
- the general background is important though
- Haupt and Haput Chapter 2
- The main point here is the binary GA, which is summarised as Matlab code at the end of the chapter
- There are many variants of each constituent step, including cross-over, matchmaking, and mutation
Briefing
- Optimisation is a general problem
- intelligent agents optimise actions
- functions are optimised in all kinds of decision making
- Two classic approaches
- exhaustive approaches - exploritng the entire solution space is rarely tractible
- iteratively improving a tentative solution is easily stuck in local maxima
- Population based methods
- multiple tentative solutions scattered over the search space
- many variants
- Most basic - look up R&N
- Genetic algorithms