### Table 2 Two-Stage Query-Based Learning Results

"... In PAGE 4: ... The best network architecture was found to contain one hidden layer with 5 neurons. QBL was found effective when we started with only a small and sparse data set as shown in Table2 . A larger training set of 2869 samples was also generated using the sampling rate of 2 along the pitch and roll angles, and 3 samples along the other parameters.... ..."

### Table 2. The performance of the two-stage (pruning and scoring) algorithm using accurate and noisy data

"... In PAGE 5: ... Our scoring takes advantage of this phenomenon, which is highly significant in loops of lengths two through five, but is still substantially noticeable in loops of lengths up to seven. 4 IMPLEMENTATION We verified the scoring function by applying it to BK TM protein chains, whose structures were solved using X-ray crystallography ( Table2 ). We restricted our study to those chains, whose TM segments did not contain half-helices or loops (except for the glycerol facilitator, as discussed below).... In PAGE 5: ... The exact definition of these regions is provided in Appendix A. The two implemented cases (using accurate and noisy data) are examined in Table2 by the number of feasible assignments that remain after the pruning phase and by the rank of the score of the native assignment with respect to other assignments. In most of the examined TM proteins, the table shows that the native assignment ranks very highly, which implies that the combination of the pruning and scoring phases yields a reliable tool for assigning TM segments to helices.... ..."

### Table 1: Two-Stage LCGs Statistics

"... In PAGE 6: ... Here after a detailed example of a PRNG in the TSRG family whose randomizer is LCG with modulus 5 and multiplier 3 with initial condition 1 denoted as a164 a11a87a101a18a14a61a165a18a14a22a33a35a24 and adapted Lehmer generator a164 a11a149a166a57a14a100a101a4a24 . Table1 summarizes the results, which imply that 3 is the total average and 2 is the bad initial condition and TSRG period a130 a7a167a11a87a168 a111 a113a54a24a36a108a95a60a49a7a104a33a38a60 .... ..."

### Table 1: Two-Stage LCGs Statistics

"... In PAGE 6: ... Here after a detailed example of a PRNG in the TSRG family whose randomizer is LCG with modulus 5 and multiplier 3 with initial condition 1 denoted as C4B4BHBN BFBN BDB5 and adapted Lehmer generator C4B4BJBN BHB5. Table1 summarizes the results, which imply that 3 is the total average and 2 is the bad initial condition and TSRG period D4 BP B4BI A2 BGB5BPBE BP BDBE.... ..."

### Table 1: Comparison of performances of two-stage and single-stage SVM approaches in protein sec- ondary structure prediction on RS126 dataset with multiple sequence alignments and the following reduction: H, G to (H); E, B to (E); the remainder to (C).

2003

"... In PAGE 8: ... Segment overlap measure (Sov) gives accuracy by counting predicted and observed segments, and measuring their overlap [3]. Table1 shows the performance of the di erent secondary structure predictors using two-level SVMs on the RS126 set with multiple sequence alignments. At the rst stage, the multi-class techniques of Vapnik and Weston gave the best result for PSS prediction which achieved 70.... ..."

Cited by 4

### Table 8: Two-stage Algorithm Test Results

2000

"... In PAGE 5: ... 4.3 Testing the Two-stage Algorithm Table8 shows the results of the Two-stage algorithm for our data sets. The maximally effective cut of point for all sets lies closer.... ..."

Cited by 6

### Table 7: Two-stage opamp macromodeling results: gain response

1996

"... In PAGE 21: ... For example, for an opamp such a macromodel would monitor the actual designed voltage gain corresponding to the input voltage gain speci cations. Such a macromodel for a two-stage opamp was developed and is shown in Table7 . In this table the rst column corresponds to the minimum input gain speci cation provided to the synthesis tool.... ..."

Cited by 10

### Table 3: Summary of results from using the two-stage approach and the

"... In PAGE 11: ...he cost for grade C is 0). Thus the total quality cost F = Q + C = 19:35. Unless otherwise stated, the unit for Q is million yen /year for the rest of the paper. The results of the initial design are summarized in Table3 . The same table also contains results given by Taguchi apos;s two-stage approach as performed by Mori (4).... ..."

### Table 1: Design specifications for two-stage op-amp.

1998

"... In PAGE 5: ... The load capacitance is 5pF and the supply voltages are Vdd = 5V and Vss = 0V. A simple design example Table1 describes the sample design problem, and shows the perfor- mance of the design obtained by GPCAD using GP1 models, and the simulated performance with BSIM1 models (HSPICE level 13). The objective was to maximize the unity gain bandwidth subject to the other given constraints.... ..."

Cited by 19

### Table 5: Two-stage opamp variable signi cance

1996

"... In PAGE 19: ... We, therefore, only have seven input variables. Using a 27?4 III design plan, we obtain the values for vi, dh i , and dl i as shown in Table5 . We use the signi cance criterion developed in subsection 3.... ..."

Cited by 10