While coevolution has many parallels to natural evolution, methods other than those based on evolutionary principles may be used in the interactive fitness setting. In this paper we present a generalization of coevolution to co-optimization which allows arbitrary black-box function optimization techniques to be used in a coevolutionary like manner.
We find that the co-optimization versions of gradient ascent and simulated annealing are capable of outperforming the canonical coevolutionary algorithm. We also hypothesize that techniques which employ non-population based selection mechanisms are less sensitive to disengagement.
[1]
Edwin D. de Jong,et al.
Ideal Evaluation from Coevolution
,
2004,
Evolutionary Computation.
[2]
W. Daniel Hillis,et al.
Co-evolving parasites improve simulated evolution as an optimization procedure
,
1990
.
[3]
John Cartlidge,et al.
Rules of engagement : competitive coevolutionary dynamics in computational systems
,
2004
.
[4]
Richard K. Belew,et al.
Coevolutionary search among adversaries
,
1997
.
[5]
Jordan B. Pollack,et al.
Co-Evolution in the Successful Learning of Backgammon Strategy
,
1998,
Machine Learning.