### Table 2. Nontrivial zeros wk and multiplicities dk for the Riemann zeta function. k

1999

### Table 1: Summary of zero dynamics analysis for the underactuated vehicle. Outputs Output Values Stability of Zero Dynamics

"... In PAGE 10: ... We expect that examining the general zero dynamics for the system in both the body- xed frame and the inertial frame will provide some insight into the motion of the underactuated vehicle and will allow us to develop higher performance controllers. As a summary of this section, we present Table1 which contains the results of our zero dynamics analysis in Sections 2 and 3.... ..."

### Table 2: The analysis of variance procedure (ANOVA) of the Q3 for the five reduction methods*

"... In PAGE 4: ...0 20 40 60 80 100 0 50 100 150 200 250 Method I lt; - ------------- N u m b e r o f p r o t e i n s --------------- gt; 0 20 40 60 80 100 0 50 100 150 200 250 Method II 0 20 40 60 80 100 0 100 200 300 Method III 0 20 40 60 80 100 0 100 200 300 Method IV 0 20 40 60 80 100 0 50 100 150 200 250 Method V lt;------------------------- Q3 % ------------------------- gt; Figure 3: Five histograms showing the Q3 distribution of the test proteins with respect to the five reduction methods Table2 shows the results of one way analysis of variance procedure (ANOVA) against the performance of prediction accuracy (Q3) of the five reduction methods. The ANOVA procedure tests for the hypothesis that whether ever all means of the five methods are similar or whether there are significant differences between them.... In PAGE 5: ... Table2 presents the results of the five reduction methods. It shows that the means are significantly different from each others at the 0.... ..."

### Table 5. Special zeros of N(s) found with the \almost period quot; method

"... In PAGE 22: ... However, with the (non-exhaustive) method described in the next section, we were able to nd a special zero of 22 near t=558,159,406 (cf. Table5 below), but this left us with the question whether that zero is the smallest special zero of 22. Only recently, the rst author suc- ceeded to nd the smallest special zero of 22 with the systematic method, in a computation which took about 105 CPU-hours on an SGI workstation.... In PAGE 24: ... In [76], we have applied two al- gorithms to nd rational approximations of log pj= log pj0, (j = 1; 2; : : :; (N), j 6 = j0), namely the modi ed Jacobi-Perron [9] and the Szekeres algorithm [131], the latter of which turned out to be more e cient than the former. We carried out various experiments, and in Table5 we present the special zeros of N(s) (rounded as in Table 4) which we found for those values of N for which we could not nd special zeros by means of the systematic method of... ..."

### Table 1: Running time of TMC and MC method. Dimension Zero

2006

Cited by 1

### Table 1. Comparison of the methods using average recall and precision values and the number of words that can be predicted with non-zero values.

2004

"... In PAGE 4: ... We prefer a method which has fewer non-used words, since it could generalize better to unseen images. Table1 shows that the proposed methods predict more words with non- zero recall and precision values (about three times more than the EM method on average). We observe that EM captions the frequent words with high precision and recall, but misses many words compare to SvdCorr/SvdCos.... ..."

Cited by 12

### Table 3: Dynamic percentages of different operand stack sizes for Java methods.

1999

"... In PAGE 3: ... 60% of method calls need a local variable array size of two or less. Table3 shows dynamic percentages of different operand stack sizes for Java methods for the programs studied. The high percentage of Java methods with zero operand stack size gives concern, as nop is the only bytecode that does not need an operand stack, but this can be explained by two factors.... ..."

Cited by 3

### Table 1: Number of non-zeros in matrix

"... In PAGE 4: ... The conjugate gradient methods turn out to have overhead per iteration in CG in relation to the rate of convergence. Moreover, the normal equation matrix surprisingly has less non-zero entries than the original matrix as show in Table1 below. This is due to the tensor product of the B-spline and the second order partial derivative that are within the profile of the B-spline subdivision matrix.... ..."

### Table 3: Dynamic method redundancy.

1998

"... In PAGE 5: ...http://www.crhc.uiuc.edu/IMPACT/). The results of our study are presented at a very high level in Table3 . The first eight rows summarize the JIT requests that occurred while running each of the SPECjvm98 benchmark applications in test mode with the largest input (control-set 100).... In PAGE 6: ...Table3 shows the number of methods requested by the JIT that were unique using byte-for- byte equivalence. As pointed out previously, a byte-for-byte equivalence does not guarantee that the methods are identical due to differences in constant pool resolution.... ..."

Cited by 1