### TABLE I FINITE DIFFERENCE GRADIENT ESTIMATOR.

2006

Cited by 9

### Table 1. Numberof responseevaluationsneededfor the gradient estimation

### TABLE I FINITE DIFFERENCE GRADIENT ESTIMATOR.

### Table 8 Tuning parameters of gradient estimation algorithm for Case Study 3

2006

"... In PAGE 16: ...3.5 Gradient estimation In Table8 the tuning parameters of the gradi- ent estimation are listed. In particular, one should note the scaling factor of the standard deviation set to 1%.... In PAGE 19: ....F. Pellissetti et al. present case np = 10). This parallel version of the algorithm a ects in par- ticular the role of the third tuning parameter N in Table8 . Here, the decision whether to start a sample pool upgrade is taken after the batch of individual gradient components is returned and not as soon as the accumu- lated number of failures exceeds the threshold.... ..."

Cited by 1

### Table 10 Tuning parameters of gradient estimation algorithm for case study 4

2006

"... In PAGE 24: ...4.3 Gradient estimation In Table10 the tuning parameters of the gradi- ent estimation are listed. For the present analysis a larger number of initial samples has been speci ed, in view of the limited CPU time required by a single response evaluation (approximately 6 seconds on an Intel Pentium,... In PAGE 26: ... The next seven evaluated components add nothing to the accuracy, before another important component is obviously identi ed, bringing the accuracy to about 73%. Thereafter, four more \fail- ures quot; are observed, which trigger an upgrade of the sample pool, since the accumulated number of failures is eleven and exceeds the value speci ed in Table10 . The remaining part of the iteration history shows a rather slow but steady convergence to the required accuracy of 95%.... ..."

Cited by 1

### Table 1. Number of response evaluations needed for the gradient estimation

### TABLE III SPECIALIZED LIKELIHOOD RATIO POLICY GRADIENT ESTIMATOR G(PO)MDP /POLICY GRADIENT WITH AN OPTIMAL BASELINE.

2006

Cited by 9

### TABLE III SPECIALIZED LIKELIHOOD RATIO POLICY GRADIENT ESTIMATOR G(PO)MDP /POLICY GRADIENT WITH AN OPTIMAL BASELINE.

### TABLE 1. Comparison of variance-based measures of opportunity for sexual selection and actual Bateman gradient estimates.*

### Table 2: Comparison of PA/FD gradient estimators PC Factor 1 2 3 4

"... In PAGE 5: ... WAC of 6.62 and pool size of $4,000,000. In order to estimate the accuracy of our PA estimator, we also estimate the gradient via finite differences (FD). Table2 gives the sensitivities of the MBS price to the prin- cipal component factors for each method. The sensitivities measure the percentage change in the price w.... In PAGE 5: ...easure the percentage change in the price w.r.t. a 1/100 change in the principal components factor coefficient. From Table2 we can see that the error is very small, and the 95% confidence intervals are almost the same. Thus, the accuracy of the PA estimator is comparable to ... ..."