Take a look at the new version of this page: [ beta version ]. Tell us what you think.

ABSTRACT

Everybody makes mistakes -- we all make one eventually if we just work hard enough! This is good news and bad news. We learn from mistakes but mistakes are also painful and could turn out to be costly in terms of money, reputation and credibility. One is prone to make mistakes particularly with new and complex techniques with unknown or not exactly known properties. This paper talks about some of my more unfortunate experiences with evolutionary computation. The paper covers design and application mistakes as well as misperceptions in academia and industry. You can make a lot of technical mistakes in evolutionary computation. However, technical errors can be detected and rectified. Algorithms are implemented, presented and analysed by humans who also discuss and measure the impact of algorithms from their very individual perspectives. A lot of 'bugs' are actually not of a technical nature, but are human flaws. This text tries also to touch on these 'soft' aspects of evolutionary computing.

REFERENCES

• 1: D. Adams. The Hitchhiker's Guide To The Galaxy. Pan Macmillan, 1979.
• 2: Thomas Bartz-Beielstein, Experimental Research in Evolutionary Computation: The New Experimentalism (Natural Computing Series), Springer-Verlag New York, Inc., Secaucus, NJ, 2006
• 3: J. Mehnen and R. Roy. Technology Transfer: Academia to Industry. In: Studies in Computational Intelligence, chapter 12, pages 263--281. Springer, London, New York, 1988.
• 4: Robert E. Keller , Wolfgang Banzhaf , Jörn Mehnen , Klaus Weinert, CAD surface reconstruction from digitized 3D point data with a genetic programming/evolution strategy hybrid, Advances in genetic programming: volume 3, MIT Press, Cambridge, MA, 1999
• 5: K. Weinert, J. Mehnen, and M. Schneider. Evolutionary Optimization of Approximating Triangulations for Surface Reconstruction from Unstructured 3D Data In: Information Processing with Evolutionary Algorithms, From Industrial Applications to Academic Speculations, chapter 3, pages 30--42. Springer, London, New York, 2005.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.4 INFORMATION SYSTEMS APPLICATIONS
      H.4.m Miscellaneous

Additional Classification:
  J. Computer Applications
  J.m MISCELLANEOUS

General Terms:
Algorithms, Design, Human Factors

Keywords:
evolutionary computation, learning from failures