### Table 1: Minimum mean cycle and minimum cost to time ratio algorithms for a graph G with n nodes and m arcs. (W, the maximum arc weight; T, the total transit time of G; N is the product of the out-degrees of all the nodes in G.)

"... In PAGE 1: ...2 Related work There are many algorithms proposed for both MCRP and MCMP. We give a comprehensive classi cation of the fastest and the most common ones in Table1 . References to a few old algorithms can be found in [12].... In PAGE 2: ... Thus, we will focus only on MCMP in this paper. In Table1 , the polynomial and pseudopolynomial algo- rithms are respectively ordered according to their worst-case running times. Those with the same running time are pre- sented in alphabetical order of their inventors apos; names.... In PAGE 2: ... We also provide some implementational improvements for most of the algorithms. In this study, we focus on the ten leading MCM algo- rithms and the MCM versions of the MCR algorithms from Table1 , all of which are named in the table. The remain- ing algorithms in this table are not included in our study because they are very similar to the chosen ones.... ..."

### Table 1: The 23 benchmark functions used in our experimental study, where n is the dimension of the function, fmin is the minimum value of the function, and S Rn. A detailed description of all functions is given in the Appendix.

1999

"... In PAGE 4: ... Using too small a test set also has the potential risk that the algorithm is biased (optimized) towards the chosen problems, while such bias might not be useful for other problems of interest. The 23 benchmark functions are given in Table1 . More detailed description of each function is given in the appendix.... In PAGE 11: ... In order to examine the impact of di erent t values on the performance of FEP in detail, a set of experiments have been carried out on FEP using di erent t values for the Cauchy mutation. Seven benchmark functions from the three di erent groups in Table1 were used in these experiments. The setup of these experiments is exactly the same as before.... In PAGE 21: ... Figures 10 and 11 show the results of IFEP, FEP, and CEP. Table1 0: Comparison among IFEP, FEP, and CEP on functions f1; f2; f10; f11; f21; f22; f23. All results have been averaged over 50 runs, where \Mean Best quot; indicates the mean best function values found in the last generation.... In PAGE 29: ...Table1 1: Kowalik apos;s Function f15 i ai b?1 i 1 0.1957 0.... In PAGE 30: ...Table1 2: Hartman Function f19 i aij; j = 1; 2; 3 ci pij; j = 1; 2; 3 1 3 10 30 1 0.3689 0.... In PAGE 30: ...5743 0.8828 Table1 3: Hartman Function f20 i aij; j = 1; ; 6 ci pij; j = 1; ; 6 1 10 3 17 3.5 1.... In PAGE 30: ...19 Shekel apos;s Family f(x) = ? m X i=1 [(x ? ai)(x ? ai)T + ci]?1 with m = 5; 7 and 10 for f21(x), f22(x) and f23(x), respectively, 0 xj 10. Table1 4: Shekel Functions f21; f22; f23 i aij; j = 1; ; 4 ci 1 4 4 4 4 0.1 2 1 1 1 1 0.... ..."

Cited by 95

### Table 4. Architectural parameters used. In the table, RT means minimum Round-Trip time from the processor. Cycle counts are in processor cycles.

2006

"... In PAGE 8: ... We call them Sub512, Sub2k, and Sub8k. The three predictors for Superset Con and Superset Agg are as follows: y512 has the y Bloom filter of Table4 and a 512-entry Exclude cache; y2k has the same Bloom filter and a 2K Exclude cache; and n2k has the n Bloom filter of Table 4 and a 2K Exclude cache. We call the resulting predictors SupCy512, SupCy2k, and SupCn2k for the Conservative algorithm and SupAy512, SupAy2k, and SupAn2k for the Aggressive one.... In PAGE 8: ... We call them Sub512, Sub2k, and Sub8k. The three predictors for Superset Con and Superset Agg are as follows: y512 has the y Bloom filter of Table 4 and a 512-entry Exclude cache; y2k has the same Bloom filter and a 2K Exclude cache; and n2k has the n Bloom filter of Table4 and a 2K Exclude cache. We call the resulting predictors SupCy512, SupCy2k, and SupCn2k for the Conservative algorithm and SupAy512, SupAy2k, and SupAn2k for the Aggressive one.... ..."

Cited by 6

### Table 2 - The ranked cooperativities of transcription factors under different environmental conditions: The number l denotes the l-th significant cooperation among these transcription factors. In this table only cooperative activities of the cell cycle are ranked.

2007

"... In PAGE 9: ...8 which are based on the statistical results in Figure 1, and (2) the common transcriptional activators in Table 1. Furthermore, the detected interactive activities among these TFs are presented in Table2 . In this study, we focus on detecting the stress-specific TFs and the common transcriptional activators that are always activative in the gene transcription process even in the absence of any specific stress; these common TFs can also be easily found by the conventional statistical method.... In PAGE 9: ... For example, our proposed method can easily find the common TFs Abf1, Rap1, Cin5, Fhl1 and Reb1 [27-31] in osmotic shock, heat shock, hydrogen peroxide treatment and cell cycle in Table 1. The interactive activities of these TFs under different environmental conditions are ranked in interactive activities matrices in Table2 . In addition, our method also can order the relative roles of the TFs in stress-specific genes of the transcriptional regulatory system.... In PAGE 13: ... The interaction between Fkh2 and Ndd1 has the second strongest regulatory ability according to our results. Furthermore, we also find strong interactivities between Swi4 and Swi6, and between Mbp1 and Swi6 ( Table2 ). According to the conventional results in the yeast cell cycle, complexes of Swi4 and Swi6 (SBF) as well as Mbp1 and Swi6 (MBF), both of which are heterodimers, are active during the G1/S phase [47,51].... In PAGE 27: ...e., 10 m = (see Table2 in the cell cycle case). In this study, for the convenience of table listing, we choose s =15 and m =10 in Equation (11) and Equation (14), respectively.... ..."

### Table 2 Comparison between calculated and experimentally determined cycle statistics for the cycle

"... In PAGE 10: ... The standard deviations were roughly estimated from available test data for various kinds of steel. Table2 summarizes the calculated and experimentally obtained statistics for the number of cycles until the onset of instable crack growth. The results are in good agreement, especially the trend in the mean value and the standard deviations for the... ..."

### Table 5 Response Surface Fits for Cycle Experimentation

"... In PAGE 7: ...9 For each of the responses monitored during the cycle experimentation (the cycle responses as well as the configuration responses for fitting the second level of these partitioned response surfaces), the mean and standard deviation data are calculated for each run of the inner control array across the runs of the outer noise array. Response surface models for mean and standard deviation are then fit to this data; resulting model fits are summarized in Table5 . For each mean and standard deviation response, second and third order response surfaces are fit, and the best fit is chosen (the modified composite experiment is a five level experiment, and thus the third order terms can be added to the basic model of Equation 3 or Equations 4-7; three-way interaction terms are not added).... In PAGE 7: ... For each mean and standard deviation response, second and third order response surfaces are fit, and the best fit is chosen (the modified composite experiment is a five level experiment, and thus the third order terms can be added to the basic model of Equation 3 or Equations 4-7; three-way interaction terms are not added). The order of fit and R2 values for each mean and standard deviation response are given in Table5 . Recall that the response models fit for the configuration mean responses in the cycle factors are actually the second portion of these models, the intercept term models (see Figure 3); the primary models for the configuration... In PAGE 8: ... With these relatively small standard deviations these standard deviation approximations for the configuration responses are accepted. In Figure 5 the response model fits for the SFC responses (mean and standard deviation) of Table5 are shown graphically as actual response data (experiment points) versus predicted response values. In these plots, the angled line represents the ideal fit (actual and predicted values being equal) around which the predicted data is scattered; the horizontal dashed line represents the response mean value.... In PAGE 8: ... Plots similar to those of Figure 5 are presented in Ref. 18 for the remaining responses of Table5 . The bands in these plots are very tight for most of the response fits.... ..."

### Table 1: Test functions used in our experimental study: the parameter n represents the problem size and fmin is the minimum value of the function. All the functions should be minimized.

2007

"... In PAGE 7: ...n the 25th generation and to 5.05 in the 85th GA generation. Starting with the 85th generation, the mean of the swarm size is stabilized at level 5 - this value seems to be the optimal size of the swarm for the considered problem. The same experiment was repeated for each test function presented in Table1 (for n = 5). The results are depicted in Figure 4.... ..."

### Table 1. Mean values of experimental results after 10 executions for each algorithm.

"... In PAGE 3: ...his task was done with the statistical package S.P.S.S. release 9.00. The results are shown in Table1 . A study of differences between particular algorithms was also done.... ..."

### Table 1: The 23 test functions used in our experimental studies, where n is the dimension of the function, fmin is the minimum value of the function, and S Rn. The detailed description of each function is given in the appendix. Test function n S

1997

"... In PAGE 4: ... This relatively large set is necessary in order to reduce biases in evaluating algorithms. The 23 test functions are listed in Table1 . The detailed description of each function is given in the appendix.... In PAGE 6: ...initial population was generated uniformly at random in the ranges speci ed in Table1 . The number of generations for each function was determined after some limited preliminary runs which showed that an ES would have converged (either prematurely or globally) after certain number of generations.... In PAGE 14: ...4 Related Work on Fast Evolutionary Programming Similar to FES, fast evolutionary programming (FEP) [8, 9] also uses Cauchy mutation. FEP has been tested on the same 23 benchmark functions as described by Table1 . Comparing those results [8, 9] with the results obtained from the current study, it is clear that the di erence between FES and CES is very similar to the di erence between FEP and CEP.... In PAGE 18: ...Table1 0: Comparison between CES and FES with recombination (discrete recombination on the objective variables and global intermediate recombination on the strategy parameters for CES, and discrete recom- bination on both objective variables and strategy parameters for FES), when a di erent scale parameter is used.... In PAGE 21: ...Table1 1: Kowalik apos;s Function f15 i ai b?1 i 1 0.1957 0.... In PAGE 22: ...Table1 2: Hartman Function f19 i aij; j = 1; 2; 3 ci pij; j = 1; 2; 3 1 3 10 30 1 0.3689 0.... In PAGE 22: ...5743 0.8828 Table1 3: Hartman Function f20 i aij; j = 1; ; 6 ci pij; j = 1; ; 6 1 10 3 17 3.5 1.... In PAGE 22: ...19 Shekel apos;s Family f(x) = ? m Xi=1[(x ? ai)(x ? ai)T + ci]?1 with m = 5; 710 for f21(x), f22(x) and f23(x), respectively, 0 xj 10. Table1 4: Shekel Functions f21; f22; f23 i aij; j = 1; ; 4 ci 1 4 4 4 4 0.1 2 1 1 1 1 0.... ..."

Cited by 26

### Table 3.1. Learning speed studies, showing the minimum, maximum, and mean number of trial iterations for the study presented in Study 2.

Cited by 1