### Table 1 depicts results from using a direct evaluation based on the product form representation, the complete partial fraction representation (CPF), and IPF representations obtained by Leja ordering for several values of . We note that the complete PF is not able to retain any correct digits. Choosing to bound the coefficients of the IPF provides good accuracy for a small number of factors m, thus preserving reasonable levels of parallelism for systems with moderate numbers of processors. For a full discussion we refer to [1].

1993

"... In PAGE 6: ...5 function evaluation method factors m value h1 CPF 1 1 :5000E0 IPF 1E9 2 :5870E ? 5 IPF 1E5 4 :5864E ? 5 DIRECT 0 40 :5865E ? 5 h2 CPF 1 1 ?:2199E + 13 IPF 1E24 2 :4143E + 10 IPF 1E10 3 :2356E + 9 DIRECT 0 40 :2356E + 9 Table1 : Examples from the evaluation of functions h1 and h2 using product form (DIRECT) and [I][PF] methods for several values of .... ..."

Cited by 1

### Table 2: Coe cients k and k computed with the the Stieltjes algorithm and 60 digits accuracy in Maple.

"... In PAGE 8: ...ion 3.1. However this process becomes very quickly unstable as k increases. Table2 shows the results of this algorithm which required 60 digits of accuracy in Maple and several hours computing time on a workstation to obtain 14 correct digits for k = 50. Table 3 shows the results of the Stieltjes algorithm with standard double precision for the example measure (4.... In PAGE 8: ... Table 3 shows the results of the Stieltjes algorithm with standard double precision for the example measure (4.1) and compares the re- sults with the accurate results obtained by Maple in Table2 . Clearly the algorithm becomes unstable and there are only one resp.... ..."

### Table 5: Dependent Variables and Their Components Dependent Variable Definition Formula

in The Effect of Communication Style on Task Performance and Mental Workload Using Wearable Computers

2000

"... In PAGE 8: ...able 4. Physical Task Counterbalanced Across Subjects............................................... 27 Table5 : Dependent Variables and Their Components .... In PAGE 27: ... Experimental Design Version Command Line Text Menus Icon Text Menus Buttons Icon Buttons Wearable Computer Off S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Assembly Wearable Computer On S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Wearable Computer Off S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Digit Wearable Computer On S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Wearable Computer Off S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Physical Task Path Walking Wearable Computer On S1-S6 S7-S12 S13-S18 S19-S24 S25-S30 Table 4. Physical Task Counterbalanced Across Subjects Task Order 1 2 3 S1 Assembly Digits Path Walking S2 Assembly Path Walking Digits S3 Digits Assembly Path Walking S4 Digits Path Walking Assembly S5 Path Walking Digits Assembly S ubj ect s (n=6) S6 Path Walking Assembly Digits Several time and accuracy measures were employed for each of the physical tasks ( Table5 ). For the assembly task, average assembly rate was used to measure the speed of assembly and was defined as the number of assembled blocks (correct or not) divided by the ... ..."

Cited by 1

### Table 8: kFk2 and cond(X) for the stabilization method. A comparison of Table 8 and Table 5 shows that the conditioning of the closed-loop system via the stabilization method is much better. It should be pointed out that in the case n = 20, the resulting closed-loop eigenvalues via the pole placement algorithm had no correct digits but those via the stabilization method had 7 valid digits. For further results in this direction, in particular for large sytems of several hundred states, see [12]. In this section we have demonstrated that for the problem of stabiliza- tion, the pole placement problem should not be considered as a substitute problem. A much better substitute (though not perfect) is the stabilization via the solution of a linear-quadratic control problem. In a similar way, one

1995

"... In PAGE 27: ... To demonstrate the superiority of this approach consider the stabilization algorithm proposed in [12] applied to one of the previous examples. Example 8 For Example 6 in the case of = 1, the result obtained from the stabilization algorithm proposed in [12] is given in Table8 . Observe that the stabilization method and the pole placement method are comparable in this case, since the eigenvalues of the closed-loop systems are both ?n; ?(n?... ..."

Cited by 1

### Table 3. % correct for several mixtures

### Table 1: Examples from the evaluation of functions h1 and h2 using product form (DIRECT) and [I][PF] methods for several values of .

1993

"... In PAGE 6: ... Let h2 := 1 p, where p(t) = Q40 k=1(t ? tk), with tk := k 40, and evaluate it at t = 1:2. Table1 depicts results from using a direct evaluation based on the product form representation, the complete partial fraction representation (CPF), and IPF representations obtained by Leja ordering for several values of . We note that the complete PF is not able to retain any correct digits.... ..."

Cited by 1

### Table 1: Results of the classi#0Ccation of handwritten digits for several sizes and

"... In PAGE 5: ... It turned out that the computational cost to calculate the tanh values can be neglected. The corresponding results are shown in Table1 for several sizes and o#0Bsets of the receptive #0Celds. Table 1 shows that the computational cost of the preprocessing procedure can be reduced by about a factor of 4 or 5 compared to the global PCA approachat a comparable computational cost of the classi#0Ccation itself without an increase of the recognition error.... In PAGE 5: ... The corresponding results are shown in Table 1 for several sizes and o#0Bsets of the receptive #0Celds. Table1 shows that the computational cost of the preprocessing procedure can be reduced by about a factor of 4 or 5 compared to the global PCA approachat a comparable computational cost of the classi#0Ccation itself without an increase of the recognition error. 3.... ..."

### Table 1. The sequences spoken throughout the experiments and their corresponding utterance number as used in the figures. Note that the same sequence is said several times throughout the recordings.

"... In PAGE 1: ... The SV system uses whole word left-right-HMMs with 2 states per phoneme and 2 mixtures per state. The system uses 4-digit sequences as passwords and the sequences used are listed in Table1 . Utterances one to four were repeated several times during the recording of the database, while utterances five to nineteen only occur once or twice per client.... In PAGE 2: ...recorded digits This experiment was set up under the assumption that it would be easy for an impostor to record the genuine client when uttering digits zero to nine and then just cut and paste these recordings to form the correct sequence during verification. The 19 sequences from Table1 were created from isolated digits uttered by the clients. This speech was then fed to the SV system in order to cause an acceptance.... ..."

### Table 5.1 level N correct digits

1982

Cited by 41

### Table 6.2 Number of correct digits of summation routines

2005

Cited by 15