Attached are my notes from the lecture.  The copies are from the
Reeves & Rowe book "Genetic Aalgorithms- Principles and Perspectives"

Other items for the students:

Frank Neumann, Dirk Sudholt and Carsten Witt (2007):
Comparing Variants of MMAS ACO Algorithms on Pseudo-Boolean Functions
http://ls2-www.cs.uni-dortmund.de/~witt/ci-23007.pdf

Benjamin Doerr, Frank Neumann, Dirk Sudholt and Carsten Witt (2007):
On the Runtime Analysis of the 1-ANT ACO Algorithm
http://ls2-www.cs.uni-dortmund.de/~witt/ci-22307.pdf  - expanded preprint
http://portal.acm.org/citation.cfm?id=1276964  - publication

Frank Neumann and Carsten Witt (2006):
Runtime Analysis of a Simple Ant Colony Optimization Algorithm
http://eccc.hpi-web.de/eccc-reports/2006/TR06-084/Paper.pdf
(to appear in Algorithmica)

Points to take away:
  0) Evolutionary Computation belongs in the family of "metaheuristic
optimization algorithms"
  1) Ant systems live in the family of 'Evolutionary Computation"
  2) It's a 30-40 year old field - but still very non-unified,
fragmented and adolescent
  3) All EC algorithms can be modeled with a the Vose Dynamic systems framework
  4) As the population size increases, all EC type algorithms will
behave according to the Vose model
  5) Some algorithms are provable the same for small population sizes
for certain families of fitness functions
  6) The No Free Lunch thoerom should serve as a warning to those
trying to claim that their heuristic model building algorithms
      is better than algorithm X in general.
  7) It's a open field (EC Theory) of interesting questions..
unfortunately there is little funding avail for this type of work.