### Table 2. Statistics about sets Si sizes on sparse random graphs with p = 4=n and dense random graphs for p = 0:3 and 0:5, at the moment we calculate cycle Ci.

2005

"... In PAGE 8: ... Instead of (nm) shortest path computations we hope to perform much less. In Table2 we study the sizes of the sets Si for i = 1; : : : ; N used to calculate the cycles for sparse and dense graphs respectively. In both sparse and dense graphs although the maximum set can have quite large cardinality, the average set size is much less than n.... ..."

Cited by 5

### Table II. Statistics about Sets Si Sizes on Sparse Random Graphs with p = 4/n and Dense Random Graphs for p = 0.3 and 0.5a

2005

Cited by 5

### Table A.2: Results on the D and E-instances. Type: Sparse random with varying graph parameters, OR-Library.

### Table 5.4: Statistics about sets Si sizes on sparse random graphs with p = 4/n and dense random graphs for p = 0.3 and 0.5, at the mo- ment we calculate cycle Ci with Algorithm 3.1. Compare the average cardinality of Si with n.

2006

Cited by 1

### Table 4: Total running times and iteration counts on sparse random graphs (cf. Sect. 4) on a 32-node CM5 of the interior-point algorithm of Sect. 2 and the first derivative method described in this section. interior-point gradient-descent

"... In PAGE 21: ... In this figure, the term processor refers to a CM5 vector unit. Table4 compares the numbers of iterations and total running times on sparse graphs of the interior- point method of Sect. 2 and the algorithm described in this section.... ..."

### Table 4.4 Comparisons with known upper bounds for max cut. These are sparse random graphs Rn;10n , data from [HP95]. The three heuristics are run for one hour of Sparcstation-20 time each. The approximation algorithm results were obtained in a matter of minutes on the CM-5.

1999

Cited by 15

### Table 7. rtc vs. the sparse method in random graphs with difierent number of vertices. The table shows for each number of vertices, the average number of constraints and the average time it took each SAT solver to solve the corresponding formula.

2005

"... In PAGE 28: ... Tables 7 and 8 present another two sets of experiments. In Table7 , the number of vertices n ranges from 200 to 500. All the instances were satisflable and the ratio between solid and dashed edges was 1:1.... In PAGE 28: ... Table 8 presents results of unsatisflable instances with n = 150 and 500 to 3000 edges, where each result again averages over 16 difierent random formulas. The advantage of rtc over sparse is clear in both Table7 and in Table 8. As was discussed in Section 7, the advantage of rtc, shown in Table 7 is expected.... ..."

Cited by 3

### Table 4.5: Asymptotic behavior of the implementations on sparse regular random graphs. (These fits do not include the unusual cases n = 1000, d = 256; 512.)

1997

Cited by 5