### Table 1 The eight possible lateral homogeneous boundary conditions and their engineering notation

"... In PAGE 3: ... The Neumann conditions are Tb 0 for each index b 2 B . Table1 shows the eight possible lateral boundary conditions. To each boundary condition we associate a space of kinematically admissible displacements: V Oe n u 2 H 1 Oe 3 ; ua 0 on @OLe 8a 2 A o : Then, the weak formulation of the elasticity problem in the three-dimensional domain is stated: Seek ue 2 V Oe , such that Oe Ae ue : e u Oe fe u, 8u 2 V Oe , 8 gt; lt; gt; : 2:1 where fe represents the volume distributed force.... ..."

### TABLE XII IRON ELASTIC REMOVAL CROSS-SECTION COMPARISON ON THE zPR-6-7 INFINITE HOMOGENEOUS MEDIUM

1977

### Table 2: Classification accuracies (%) of various dimensional spaces.

2007

"... In PAGE 14: ...thers. Similar observation was also made in previous studies, e.g. in [35]. Therefore, in this experiment the 1-NN classifier is used. Table2 shows the classification accuracy results for different dimensions of feature spaces on the ORL data sets. The accuracy of GLLE is 95.... ..."

### Table 2: Benchmark Characteristics. Copying and colocation are measured using a 4 MB nursery and infinite mature space.

"... In PAGE 7: ... 7.1 Potential of colocation Table2 presents the allocation characteristics of our benchmarks: the total allocation in MB (Total) and the amount the collector pro- motes from the nursery to the mature space (Copy) in MB and as a percentage. We order programs by their nursery survival rate.... In PAGE 8: ... To focus on accuracy, we measure these values using specially instrumented collectors configured with a 4 MB nursery with an in- finite mature space. ( Table2 also presents these raw numbers and adds percentages.) Each bar shows the amount of memory that ends up in the mature space, broken down into two parts: the dark part represents memory copied from the nursery and the light part represents memory allocated directly in the mature space.... In PAGE 8: ...4 Write barrier Colocation also has the potential to reduce intergenerational point- ers, and therefore reduce the number of write barriers. The last two columns of Table2 show the percent of all writes that the write barrier records in the remembered set (remset). We observe this secondary reduction for pseudojbb, javac, and db, which is not surprising since these are the benchmarks for which colocation is most effective.... ..."

### Table I. Two-dimensional analysis of a semi-infinite crack in a linear elastic solid subjected to mode I loading.

### Table 1: Initial evaluation of cost of chaotic dynamic place- ment. Here an infinitely connected idealized placement is compared against our dynamic placement algorithms. The numbers represents the maximum available IPC given a par- ticular scheduling algorithm.

"... In PAGE 3: ... However, we can make a highly pes- simistic assessment by comparing our placement algorithms to a completely connected mesh of computation, so instruc- tions always receive data from adjacent tiles (think of this as a computational cache existing in a highly-dimensional physical space). Table1 illustrates our results for a sub- set of the SPEC2000 benchmark suite. The data demon- strate that choosing the best placement algorithm for an ap- plication yields a 57% performance penalty from an ideal schedule.... ..."

### Table 9: Experiments using 3 dimensional feature space on Dataset 4

"... In PAGE 6: ... We need some algorithm that eliminates automatically such outliers. We compare accuracy values between 2 and 3 dimen- sional feature spaces (see Table9 ). The convergence size Table 9: Experiments using 3 dimensional feature space on Dataset 4... ..."

### Table 1: Algorithm for planning in low-dimensional belief space.

in Abstract

"... In PAGE 4: ... Our conversion algorithm is a variant of the Augmented MDP, or Coastal Navigation algorithm [9], using belief features instead of entropy. Table1 outlines the steps of this... ..."