### Table 7: % WER on eval98 for the evaluation system and a com- pletely separate MLE model-based (b) branch (pure MLE).

2000

### Table 2: Alignment Accuracy Results on a Model- based Dataset.

"... In PAGE 5: ... We ob- served similar results as seen in T:B1 for the template-based dataset. In Table2 we report only the best results achieved rather than showing results for varying a2 a18 a19 a20 sizes as done in Table 1. Firstly, we notice the difference in the LGscore values for the family, superfamily and fold pairs clearly showing the dif- ficulty nature of the three sets of problems, with the fold-pairs being the hardest to model followed by the superfamily and family level pairs.... ..."

Cited by 1

### Table 2: Alignment Accuracy Results on a Model- based Dataset.

"... In PAGE 6: ... We ob- served similar results as seen in T:B1 for the template-based dataset. In Table2 we report only the best results achieved rather than showing results for varying a2a18a19a20sizes as done in Table 1. Firstly, we notice the difference in the LGscore values for the family, superfamily and fold pairs clearly showing the dif- ficulty nature of the three sets of problems, with the fold-pairs being the hardest to model followed by the superfamily and family level pairs.... ..."

Cited by 1

### Table 8. Summary of analysis for models based on two class variables CLASS

1996

"... In PAGE 21: ... Only combinations of two class variables at a time were considered; categorizing the data by three class variables led to too many cells with only one observation. Table8 summarizes the four best models... ..."

### Table 1. Distribution model based on InSAR information

2000

"... In PAGE 6: ...nside the fault. The position of the interpolating point relative to the fault can be determined in the same way. The combination of open- and closed-curve models can deal with comparatively complex fault systems. After the classification of GPS stations and the intended interpolating points into different groups, a distribution model for interpolation for each group based on the GPS-corrected InSAR results is proposed, as illustrated in Table1 . In some sites of interest (Sites 1 to 4) there are both GPS and InSAR results for the deformation (the CGPS results may have to span one or more InSAR repeat cycles depending on the availability of suitable SAR image pairs).... ..."

Cited by 1

### Table 1: Since the results of registration obtained by the observers are highly correlated, the model{based registration technique is observer{independent. Observers

"... In PAGE 3: ... The normalized cross correlation coe cients were computed to indicate the similarity of the image data after registration [13]. Table1 shows the strong correlation between the registrations obtained by the three observers. 4 Discussion Although subtraction techniques are superior in diagnosing small changes in radiographs [2, 3, 4] automatic registration algorithms are still limited to RST{movements [8] and landmark based algorithms are strongly de- pendent on the observer who positions the points [7].... ..."

### Table 5 Comparison of the calculated values of Lt with the data of [1] for two model-based estimates, an upper limit and a lower limit

1998

"... In PAGE 4: ... An upper limit is obtained because the track has to disappear when the bundle radius equals the track size: 0:4Lt=v 1:4 1:34 dE=dx 1=2; 9 so that Lt 2:5v 1:34 dE=dx 1=2 0:714; 10 where v is the velocity of the cluster. These results are given in Table5 , columns 5 and 7, where we Table 3 Data from [1] and the comparison to the relationship given by the model Lc / n0:24 E=n 1=2 Projectile Energy (MeV) Experimental Lc ( A) Lc= n0:24 E=n 1=2 rs ( A) C20 13.4 940 560 11.... In PAGE 5: ...track length by about 15%. In a similar way, we can set a lower limit on track length by expressing the minimum track ra- dius in terms of a minimum number of C atoms, ni ( Table5 , column 9), that will leave a visible track. We then use 0:4Lt=v 1:4 32:8 n=ni 1=3 11 to give the depth at which groups of ni atoms are just far apart enough to give minimum tracks.... In PAGE 5: ... We then use 0:4Lt=v 1:4 32:8 n=ni 1=3 11 to give the depth at which groups of ni atoms are just far apart enough to give minimum tracks. As shown in columns 6 and 8 of Table5 , we consis- tently underestimate the track length by about 10%. 4.... ..."

### Table 1: A summary of the model-based algorithms described in this paper. The ! column contains d if the internal state update is deterministic and s if the update is stochastic. Similarly, the column indicates if the choice of action is deterministic of stochastic. Uppercase D indicates that the function is xed instead of learnt. The last column describes how parameterises !(hj ; g; y) and parameterises (uj ; h; y).

in A (Revised) Survey of Approximate Methods for Solving Partially Observable Markov Decision Processes

2003

"... In PAGE 29: ...cales [Precup, 2000]. Ghavamzadeh and Mahadevan [2001] and Makar et al. [2001] extend MAXQ to continuous time and demonstrate the algorithm in a multi-agent automated guided vehicle setting. 9 Summary Table1 summarises the model-based algorithms described in this paper. Table 2 correspondingly summarises the model-free algorithms.... In PAGE 31: ...column has the same meaning as Table1 . The tables are not a complete sum- mary of all POMDP algorithms.... ..."

Cited by 22

### Table 9: Performance of Batch-Means Procedures for the Time-Shared Model Based on 100 Independent Repli- cations of Nominal 90% Confidence Intervals Precision Procedure

"... In PAGE 7: ... (See Law and Carson 1979 for a more complete description of the model.) Table9 summarizes the results for the time-shared computer system. For this test problem, ASAP exhibits almost ideal behavior.... ..."

### Table 1: Basic concepts in model-based diagnosis and over-determined constraint satisfaction problems. Except for observations and domain of variables the concepts in model-based diagnosis and over-determined constraint satisfaction problems are quite similar.

1993

"... In PAGE 2: ... 2.3 Analogy between OCSPs and MBD The basic concepts in model-based diagnosis and over- determined constraint satisfaction problems are summarized in Table1 . Components and constraints are quite similar, as is the notion of connections, weights, and likely diagnoses and low weighted overruled constraints.... ..."

Cited by 24