Genetic programming: on the programming of computers by means of natural selection
Genetic programming: on the programming of computers by means of natural selection
Solving the artificial ant on the Santa Fe trail problem in 20,696 fitness evaluations
Proceedings of the 9th annual conference on Genetic and evolutionary computation
The reliability of confidence intervals for computational effort comparisons
Proceedings of the 9th annual conference on Genetic and evolutionary computation
Parallel evolution using multi-chromosome cartesian genetic programming
Genetic Programming and Evolvable Machines
Genetic programming that ensures programs are original
CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Representation and structural biases in CGP
CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
Self-adaptive focusing of evolutionary effort in hierarchical genetic programming
CEC'09 Proceedings of the Eleventh conference on Congress on Evolutionary Computation
More on computational effort statistics for genetic programming
EuroGP'03 Proceedings of the 6th European conference on Genetic programming
Confidence intervals for computational effort comparisons
EuroGP'07 Proceedings of the 10th European conference on Genetic programming
The performance of a selection architecture for genetic programming
EuroGP'08 Proceedings of the 11th European conference on Genetic programming
Confidence intervals of success rates in evolutionary computation
Proceedings of the 12th annual conference on Genetic and evolutionary computation
Extending genetic programming to evolve perceptron-like learning programs
ICAISC'10 Proceedings of the 10th international conference on Artifical intelligence and soft computing: Part II
EuroGP'11 Proceedings of the 14th European conference on Genetic programming
Genetic programming needs better benchmarks
Proceedings of the 14th annual conference on Genetic and evolutionary computation
Better GP benchmarks: community survey results and proposals
Genetic Programming and Evolvable Machines
Reducing wasted evaluations in cartesian genetic programming
EuroGP'13 Proceedings of the 16th European conference on Genetic Programming
| Hi-index | 0.00 |
As research into the theory of genetic programming progresses, more effort is being placed on systematically comparing results to give an indication of the effectiveness of sundry modifications to traditional GP. The statistic that is commonly used to report the amount of computational effort to solve a particular problem with 99% probability is Koza's I(M, i, z) statistic. This paper analyzes this measure from a statistical perspective. In particular, Koza's I tends to underestimate the true computational effort, by 25% or more for commonly used GP parameters and run sizes. The magnitude of this underestimate is nonlinearly decreasing with increasing run count, leading to the possibility that published results based on few runs may in fact be unmatchable when replicated at higher resolution. Additional analysis shows that this statistic also underreports the generation at which optimal results are achieved.