### Table 1: Parameter values for Markovian experiments

2006

### Table 2: F-score, compression, number of iterations, and elapsed time for unregularized and standardized maximum-likelihood estimation, and lscript0, lscript1, and lscript2 regularization on test split of PARC 700 dependency bank.

"... In PAGE 5: ... 6.3 Experimental Results Table2 shows the results of an evaluation of five different systems of the test split of the PARC 700 dependency bank. The presented systems are unreg- ularized maximum-likelihood estimation of a log- linear model including the full feature set (mle), standardized maximum-likelihood estimation as de- scribed in Sect.... ..."

### Table 1: Estimated parameters. stands for the proportion of the two classes. and 2 are the Gaussian parameters (shadow area). min and are the Rayleigh law parameters (sea oor re- verberation) [8]. i apos;s are the a priori parameters of the Markovian SCM model.

"... In PAGE 3: ... Figures 3,4 show the segmentation results obtained with the di erent meth- ods. The MRF prior model parameters x and the noise model parameters y obtained with our scheme are given in Table1 . Experiments indicate that the SMAP requires less computation than the multigrid or SCM algorithms but the neighborhood structure is not complex enough to describe local image properties.... ..."

### Table 3: Maximum-likelihood power estimates for the in- dustrial circuit

in Optimum Probability Model Selection Using Akaike's Information Criterion For Low Power Applications

"... In PAGE 4: ... Table 2 shows the optimum value of CZD3D4D8 BP BI computed by the proposed algorithm agrees with the histogram in Fig 2. The maximum-likelihood esti- mates of the average power consumption corresponding to each of the six components as seen in the histogram is sum- marized in Table3 [5]. We note that a Monte Carlo simula- tor would have captured only the most probable component and missed the remaining five components (in this case).... ..."

### TABLE VIII Labeling accuracy under 2-state Markovian-Gilbert link error model. In the Smooth case, the time duration distribution in GOOD and BAD states is CUBZC7C7BW BP BCBMBLBLBNBUBTBW BP BCBMBCBDCV. In the Busrty case, the distribution is CUBZC7C7BW BP BCBMBLBNBUBTBW BP BCBMBDCV. Network configuration II is used in both cases.

2003

Cited by 33

### Table 2: Run time statistics of our bounding technique in the micropipeline experiments. due to the strong cyclic structure of the micropipelines, state-of-art Markovian approach (e.g., [19]) su ers from the state explosion problem and cannot handle the micropipeline models of more than 8 stages within a reasonable amount of time.

1999

"... In PAGE 16: ...4 While signi cantly better than Campos et al apos;s upper bound, they are not as sharp as ours.5 Table2 lists the number of unfoldings performed, the sample size in Monte Carlo sampling and the run time for each experiment. The sample size remains roughly the same as the number of stages increases.... ..."

Cited by 16

### TABLE VIII Labeling accuracy under 2-state Markovian-Gilbert link error model. In the Smooth case, the time duration distribution in GOOD and BAD states is CUBZC7C7BW BP BCBMBLBLBNBUBTBW BP BCBMBCBDCV. In the Busrty case, the distribution is CUBZC7C7BW BP BCBMBLBNBUBTBW BP BCBMBDCV. Network configuration II is used in both cases.

### Table 3: estimated parameterson the picture reported in Figure 7:a. stands for the proportionof the two classes within the sonar image. and 2 are the Gaussian parameters (shadow area). min and are the Rayleigh law parameters (sea oor reverberation). i apos;s are the a priori parameters of the Markovian modeling. [0] represents the initial parameter estimates and the nal estimates are denoted ^ .

"... In PAGE 12: ... The mixture of distributions is represented by Figures 7:d and 8:d and the nal result of the segmented image is reported in Figures 7:e and 8:e. The obtained results are given in Table3 and Table 4. One can see that this approach gives convincing results and allows to con- verge toward a good image segmentation in spite of speckle noise.... ..."

### Table 4: estimated parameters on the picture reported in Figure 8:a. stands for the proportionof the two classes within the sonar image. and 2 are the Gaussian parameters (shadow area). min and are the Rayleigh law parameters (sea oor reverberation). i apos;s are the a priori parameters of the Markovian modeling. [0] represents the initial parameter estimates and the nal estimates are denoted ^ .

### Table 2: Trip Generation Model Results

in and

2003

"... In PAGE 10: ...9 The trip generation model was estimated using full-information maximum likelihood code programmed in GAUSS software (Aptech 1999). The maximum-likelihood parameter estimates for this model appear in Table2 .... ..."