We present a method for estimating fitness functions that are computationally expensive for an exact evaluation. The proposed estimation method applies a number of partial evaluations based on incomplete information or uncertainties. We show how this method can yield results that are close to similar methods where fitness is measured over the entire dataset, but at a fraction of the speed or memory usage, and in a parallelizable manner. We describe our experience in applying this method to a real-world application in the form of evolving equity trading strategies.

Other

In Riolo, R., Vladislavleva, E., Ritchie, M., Moore, J., editors, Genetic Programming Theory and Practice X, University of Michigan, Ann Arbor, USA, May 2013. Springer, New York, NY..

@inproceedings{hodjat:gptp13, title={Introducing an Age-Varying Fitness Estimation Function}, author={Babak Hodjat and Hormoz Shahrzad }, booktitle={Genetic Programming Theory and Practice X}, month={May}, editor={Riolo, R., Vladislavleva, E., Ritchie, M., Moore, J.}, address={University of Michigan, Ann Arbor, USA}, publisher={Springer, New York, NY.}, url="http://nn.cs.utexas.edu/?hodjat:gptp12", year={2013} }