### Table 1. Genetic programming parameters

2004

"... In PAGE 4: ... Genetic programming was generational, with crossover and mutation similar to those outlined by Koza in [13]. The parameters used by GP are shown in Table1... ..."

Cited by 5

### Table 3: Genetic programming parameters

in Evolving Cooperative Control on Sparsely DistributedTasks for UAV Teams Without Global Communication

2007

"... In PAGE 9: ... Evolution was generational, with crossover and mutation similar to those outlined in [2]. The parameters used by GP to evolve controllers are shown in Table3 . Tournament selection was used.... ..."

### Table 3: Genetic programming parameters

2007

"... In PAGE 9: ... Evolution was generational, with crossover and mutation similar to those outlined in [2]. The parameters used by GP to evolve controllers are shown in Table3 . Tournament selection was used.... ..."

### Table 2: Summery of the Genetic Programming Parameters

2005

"... In PAGE 3: ... The players are able to adopt different strategies thus each player should have a population consisting of candidate strategies. Both populations employ the same values of genetic opera- tors: the population size, the generation size, the crossover rate, the mutation rate and the method of selection as stated in Table2 . Each population evolves following Evolution- ary Algorithms, whilst the fitness assessments are subject to the co-evolving opponent population at the concurrent evo- lutionary time.... ..."

Cited by 1

### Table 4: Program Depth, standard crossover 50 runs

"... In PAGE 12: ... It is evident that the linear growth in program mean depth is not a uke but may be an important property of standard subtree crossover (in the absence of depth or size limits). Table4 gives the mean and max program depths and their average rate of increase over the last 38 generations of the runs. While not problem independent, Table 4 shows the rate of increase in depth is consistently close to unity.... In PAGE 12: ... Table 4 gives the mean and max program depths and their average rate of increase over the last 38 generations of the runs. While not problem independent, Table4 shows the rate of increase in depth is consistently close to unity. 6.... ..."

### Table 4: Program Depth, standard crossover 50 runs

1999

"... In PAGE 14: ... It is evident that the linear growth in program mean depth is not a n0duke but may be an important property of standard subtree crossover n28in the absence of depth or size limitsn29. Table4 gives the mean and max program depths and their average rate of increase over the last 38 generations of the runs. While not problem independent, Table 4 shows the rate of increase in depth is consistently close to unity.... In PAGE 14: ... Table 4 gives the mean and max program depths and their average rate of increase over the last 38 generations of the runs. While not problem independent, Table4 shows the rate of increase in depth is consistently close to unity. 6.... ..."

### Table 2: Parameters for Genetic Algorithm

"... In PAGE 4: ... Table2 summarises the various settings for our GA to search for fitter programs. Note that we experimented with a number of values for string length, crossover and mutation rate and for the number of generations.... ..."

### Table 4.3 - Genetic crossover operations

2007

### Table 4: Program Depth, standard crossover 50 runs Problem Initiali- Final pop Growth per gen

2000

"... In PAGE 4: ... It is evident that the linear growth in program mean depth is not a uke but may be an important property of standard subtree crossover (in the absence of depth or size lim- its). Table4 gives the mean and max program depths and their average rate of increase over the last 38 gen- erations of the runs. While not problem independent, Table 4 shows the rate of increase in depth is consis- tently close to unity.... In PAGE 4: ... Table 4 gives the mean and max program depths and their average rate of increase over the last 38 gen- erations of the runs. While not problem independent, Table4 shows the rate of increase in depth is consis- tently close to unity. As discussed in [8] this, together with remaining close to the ridge in the number of programs versus their shape leads to a prediction of growth of sub-quadratic growth in program length (for modest size programs we expect size O(gens1:3) rising to a limit of quadratic growth for jprogramj 1000.... ..."

Cited by 62

### Table 4: Program Depth, standard crossover 50 runs Problem Initiali- Final pop Growth per gen

2000

"... In PAGE 4: ... It is evident that the linear growth in program mean depth is not a uke but may be an important property of standard subtree crossover (in the absence of depth or size lim- its). Table4 gives the mean and max program depths and their average rate of increase over the last 38 gen- erations of the runs. While not problem independent, Table 4 shows the rate of increase in depth is consis- tently close to unity.... In PAGE 4: ... Table 4 gives the mean and max program depths and their average rate of increase over the last 38 gen- erations of the runs. While not problem independent, Table4 shows the rate of increase in depth is consis- tently close to unity. As discussed in [8] this, together with remaining close to the ridge in the number of programs versus their shape leads to a prediction of growth of sub-quadratic growth in program length (for modest size programs we expect size O(gens1:3) rising to a limit of quadratic growth for jprogramj 1000.... ..."

Cited by 62