Last ned (Gratis)

Abstrakt

Biological development has motivated researchers to apply artificial development
in bio-inspired systems. Among the possible features of artificial development
that are being investigated is the potential for improving scalability of
evolutionary optimization techniques, by applying artificial development as
an indirect mapping. Evolutionary design optimization has proved successful
for several applications but is often not applicable on its own as a large scale
problem solver due to scalability issues. Complex, Reconfigurable, Adaptive
and Bio-inspired Hardware (CRAB) Lab has investigated the challenge
in several projects. Here we present a summary of some of these projects in
the context of complexity and scalability.

Referanser

[1] EGGENBERGER, P. Evolving morphologies of simulated 3d organisms based on
differential gene expression. In Proc. the 4th European Conference on Artificial Life
(ECAL97) (1997), Cambridge: MIT Press, pp. 205–213.
Scalability and complexity challenges in evolutionary 15
i
i
i
i
[2] GORDON, T., AND BENTLEY, P. Towards development in evolvable hardware. In
Proc. The 2002 NASA/DoD Conference on Evolvable Hardware (15-18 July 2002),
IEEE, pp. 241–250.
[3] GRUAU, F. Neural Network Synthesis using Cellular Encoding and the Genetic
Algorithm. PhD thesis, France, 1994.
[4] GRUAU, F., WHITLEY, D., AND PYEATT, L. A comparison between cellular
encoding and direct encoding for genetic neural networks. In Genetic Programming
1996: Proc. the First Annual Conference, MIT Press.
[5] HADDOW, P. C., HARTMANN, M., AND DJUPDAL, A. Addressing the metric
challenge: Evolved versus traditional fault tolerant circuits. In Second NASA/ESA
Conference on Adaptive Hardware and Systems (AHS) (2007), pp. 431–438.
[6] HARTMANN, M., AND GOEDEWEECK, T. Adapting a genotype-phenotype
mapping to phenotypic complexity. In NASA/ESA Conference on Adaptive
Hardware and Systems (AHS) (2009), pp. 35–42.
[7] HARTMANN, M., AND HADDOW, P. C. Evolution of fault-tolerant and noise-robust
digital designs. IEE Proc. -Comput. Digit. Tech. 151, 4 (2004), 287–294.
[8] HARTMANN, M., HADDOW, P. C., AND LEHRE, P. K. The genotypic complexity
of evolved fault-tolerant and noise-robust circuits. Biosystems 87, 2-3 (2007), 224–
232.
[9] HARTMANN, M., LEHRE, P. K., AND HADDOW, P. C. Evolved Digital Circuits and
Genome Complexity. In Proceedings of 2005 NASA/DoD Conference on Evolvable
Hardware (2005), IEEE Press, pp. 79–86.
[10] KITANO, H. Designing neural networks using genetic algorithm with graph
generation system. Complex Systems 4 (1990), 461–476.
[11] LEHRE, P. K., AND HARTMANN, M. Complexity-based fitness function modifiers.
In Proc. ofWorkshop on Learning and Regeneration in Developmental Systems (part
of GECCO2004) (2004).
[12] LI, M., AND VIT´ANYI, P. M. B. An Introduction to Kolmogorov Complexity and
Its Applications, 2nd ed. Springer-Verlag, New York, 1997.
[13] MILLER, J. F., JOB, D., AND VASSILEV, V. K. Principles in the evolutionary
design of digital circuits – part i. Journal of Genetic Programming and Evolvable
Machines 1, 1 (2000), 8–35.
[14] MILLER, J. F., AND THOMSON, P. A developmental method for growing graphs
and circuits. In Proc. the 5th International Conference on Evolvable Systems
(ICES03) (2003), pp. 93–104.
[15] SIDDIQI, A., AND LUCAS, S. A comparison of matrix rewriting versus direct
encoding for evolving neural networks. In Proc. the 1998 IEEE International
Conference on Evolutionary Computation (1998), IEEE Press, pp. 392–397.
[16] ZIV, J., AND LEMPEL, A. A universal algorithm for sequential data compression.
IEEE Transactions on Information Theory 23 (1977), 337–342.