Results 1 - 10 of 162,037

Table 4. Comparative Evaluation of Genetic Algorithms

in Effective Black-Box Testing with Genetic Algorithms
by Mark Last, Shay Eyal, Abraham K
"... In PAGE 10: ...3 0.3 - MinLT = Minimum lifetime (number of generations) 1 1 MaxLT = Maximum lifetime (number of generations) 7 7 - 4 Summary of Results Table4 presents a summarized comparison for each performance measure obtained with each algorithm. The results of each algorithm were averaged over 20 runs.... In PAGE 10: ... FAexGA#1 is the most effective configuration, since it found the highest number of distinct solutions in its final population. Table4 also shows that all genetic algorithms prove to be considerably more efficient for this problem than the conventional test generation methods. According to ... ..."

Table 2. Comparison of the number of evaluations required by the difierent algorithms in RND with square shaped coverage antennae

in Optimal Placement of Antennae using Metaheuristics
by Enrique Alba, Guillermo Molina, Francisco Chicano
"... In PAGE 5: ... For the CHC algorithm the parameter tuned is the population size. The results of the experiments are shown in Table2 for square shaped cell antennae and Table 3 for omnidirectional antennae. All the algorithms were able to solve the problem with very high hit ratio (percentage of executions where the optimal solution is found) except a few exceptions (highlighted in italics), therefore only the number of evaluations is shown.... ..."

Table 1. Genetic Algorithm Results: Objective Values.

in A Random-Key Genetic Algorithm for the Generalized Traveling Salesman Problem
by Lawrence V. Snyder, Mark S. Daskin 2000
"... In PAGE 15: ... 4.2 Solution Quality Table1 summarizes the results for each of the problems. The columns are as follows: Problem: The name of the test problem.... ..."
Cited by 4

Table 1. Genetic Algorithm Results: Objective Values.

in A random-key genetic algorithm for the generalized traveling salesman problem
by Lawrence V. Snyder, Mark S. Daskin 2000
"... In PAGE 15: ... 4.2 Solution Quality Table1 summarizes the results for each of the problems. The columns are as follows: Problem: The name of the test problem.... ..."
Cited by 4

Table 1: Comparison of the best networks for problem (4) found by a genetic algorithm and by sequential evaluation.

in Topology Design of Feedforward Neural Networks By Genetic Algorithms
by Slawomir W. Stepniewski, Andy J. Keane 1996
Cited by 4

Table 6. The Roster of Gentic Algorithms with Multiple Objectives for the Sequential Model

in The integrated scheduling and rostering problem of train driver using
by Chi-kang Lee
"... In PAGE 21: ... The objectives are the number of duty, the length of duty, the length of worktime, and the variance of duty worktime. As the roster result illustrated in Table6 , we solve the rostering sub-problem using a Genetic algorithm with the new solution of the scheduling sub-problem with multiple objectives. It is clear that the sequential model with multiple objectives produces the cycle length of roster for 46 days, which is as good as the result of the integrated model in Table 3.... ..."

Table 7: Comparison of CHC+, Genitor, and Genitor+ on 32 test problems. CHC+ is the CHC algorithm with thresholds reset with respect to number of model lines. Genitor uses a constant mutation rate, while Genitor+ uses the new adaptive mutation operator that is also sensitive to number of model lines. Params describes problem parameters: the number indicates the length of the string encoding, while M indicate multiple objects and C indicates clutter. Ps denotes per cent of time that a global solution is found. Evals refers to the number of evaluations.

in Test Driving Three 1995 Genetic Algorithms: New Test Functions and Geometric Matching
by D. Whitley, R. Beveridge, C. Graves, K. Mathias 1995
"... In PAGE 20: ... Thus, while CHC dominated the other algorithms on our test suite, it failed to generate competitive solutions on the geometric matching problems. Table7 illustrates the relative performance of the enhanced CHC+ algorithm, Genitor and Genitor+ using the new adaptive mutation operator. The results illustrate the relatively poor performance of CHC+ compared to Genitor.... In PAGE 21: ...The maximum likelihood estimate of this probability, Ps, is simply the percent of correct solutions found when running the genetic algorithm multiple times. These probabilities are shown in Table7 for each of the three algorithms. Table 7 also shows the the number of evaluations required to reach the stopping criterion which terminates the genetic search in the column labeled Evals.... In PAGE 21: ... These probabilities are shown in Table 7 for each of the three algorithms. Table7 also shows the the number of evaluations required to reach the stopping criterion which terminates the genetic search in the column labeled Evals. 5.... In PAGE 21: ... The second is how much work must be performed in order to reach convergence. The rst is easily measured for sample problems as illustrated above in Table7 . The second is more problematic.... ..."
Cited by 16

Table 2 Non-dominated solutions obtained by the hybrid two-objective genetic algorithm

in Selecting linguistic classification rules by twoobjektive genetic algorithms
by Hisao Ishibuchi, Tadahiko Murata 1995
"... In PAGE 6: ... Simulation Result We apply the hybrid two-objective genetic algorithm to the classification problem of the iris data. We show the obtained non-dominated solutions in Table2 . From the comparison of Table 2 with Table 1, we can see that the learning method incorporated in the two-objective genetic algorithm improved the classification performance (i.... In PAGE 6: ... We show the obtained non-dominated solutions in Table 2. From the comparison of Table2 with Table 1, we can see that the learning method incorporated in the two-objective genetic algorithm improved the classification performance (i.e.... ..."
Cited by 2

Table 1: Tableau describing the EA for the Antenna placement problem

in Evolutionary Computing in Telecommunication Network Design: A Survey
by P. Kampstra A, A. E. Eiben A

Table 1. The 50 pre-processing functions made available for evaluation by the genetic algorithm for the PLS optimisation study.

in GENETIC ALGORITHM OPTIMISATION FOR PRE-PROCESSING AND VARIABLE SELECTION OF SPECTROSCOPIC DATA
by Roger M. Jarvis, Royston Goodacre
"... In PAGE 13: ... The RMSEP is given as: = = n i pred act n y y 1 2 / ) ( RMSEP , (3) where yact are the actual dependent variables, ypred are the predicted dependent variables and n is the number of objects. A GA was used to recover subsets of pre-processing functions from a total of 50 alternatives ( Table1 ). These are by no means a comprehensive list of spectral pre- processing algorithms available to the analyst; but represent the broad range of tools which in general can be split in to scaling, filtering, baseline correction and derivatisation categories.... In PAGE 15: ...5 x 106 days computation time on a 1.6Ghz PC! In Table 2 the best results from each of the independent runs are provided with the selected pre-processing functions encoded as detailed in Table1 . For comparison, details are shown for the model constructed against the raw data and for exhaustive searches using a single function.... ..."
Results 1 - 10 of 162,037