Comparing direct and developmental encoding schemes in artificial evolution: a case study in evolving lens shapes

In this paper, different evolutionary encoding schemes were compared. In a simulator for geometrical optics different lens shapes in three dimensions were evolved using direct and indirect encodings. Direct and indirect methods were compared for their precision, convergence and efficiency. The results showed that the indirect encoding schemes converged faster than the direct ones, needed less genetic parameters and scientifically most important, it could be understood why the indirect scheme outperformed the direct one.

[1]  Peter J. Bentley,et al.  On growth, form and computers , 2003 .

[2]  Peter Eggenberger.  Cell Interactions as a Control Tool of Developmental Processes for Evolutionary Robotics , 1996 .

[3]  Peter J. Bentley,et al.  Three Ways to Grow Designs: A Comparison of Embryogenies for an Evolutionary Design Problem , 1999, GECCO.

[4]  P. Eggenberger,et al.  An evolutionary approach to pattern formation mechanisms on lepidopteran wings , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[5]  Josh Bongard,et al.  Evolving modular genetic regulatory networks , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[6]  P. E. Hotz,et al.  Evolving the morphology of a neural network for controlling a foveating retina: and its test on a real robot , 2002 .

[7]  R. Beer,et al.  20 – A developmental model for the evolution of complete autonomous agents , 2003 .

[8]  Peter Eggenberger Hotz.  Genome-physics interaction as a new concept to reduce the number of genetic parameters in artificial evolution , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[9]  Maja J. Matarić,et al.  A Developmental Model for the Evolution of Complete Autonomous Agents , 1996 .

[10]  Paulo S. Salomon.  Applying evolutionary algorithms to real-world-inspired problems with physical smoothness constraints , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[11]  Peter Eggenberger.  Creation of Neural Networks Based on Developmental and Evolutionary Principles , 1997 .

[12]  Peter Eggenberger,et al.  Evolving Morphologies of Simulated 3d Organisms Based on Differential Gene Expression , 1997 .