### Table 3: The number of cases in which the P- Loc algorithm found larger/smaller probabili- ties than DWA* in network Andes when spend- ing a little bit more time than DWA*.

"... In PAGE 6: ... However, in practice the search time is not continuous in the number of search steps, so we just chose parameters for P-Loc such that it spent only a little bit more time than DWA*. Table3 shows the compar- ison results. We can see that after increasing the search steps of P-Loc, DWA* still main- tains better accuracy.... ..."

### Table 3: The number of cases in which the P- Loc algorithm found larger/smaller probabili- ties than DWA* in network Andes when spend- ing a little bit more time than DWA*.

"... In PAGE 6: ... However, in practice the search time is not continuous in the number of search steps, so we just chose parameters for P-Loc such that it spent only a little bit more time than DWA*. Table3 shows the compar- ison results. We can see that after increasing the search steps of P-Loc, DWA* still main- tains better accuracy.... ..."

### TABLE IV THE SUM OF CONNECTION DURATIONS ASSUMING THE DOMINANT TRAFFIC DIRECTION OF EACH CONNECTION IS KNOWN OR UNKNOWN. THE COMPARISON SHOWS THE KNOWLEDGE OF TRAFFIC DIRECTION ONLY HELPS A LITTLE BIT IN DECREASING THE SUM OF CONNECTION DURATIONS.

2004

Cited by 23

### Table 5: Detailed timings for vectorized monomial operations for DEC Alpha (64 bit, little{endian) running Linux.

1998

Cited by 4

### Table 7: Detailed timings for rank-based monomial operations on a DEC Alpha (64 bit, little{endian) running Linux.

1998

Cited by 4

### TABLE IV THE SUM OF CONNECTION DURATIONS ASSUMING THE DOMINANT TRAFFIC DIRECTION OF EACH CONNECTION IS KNOWN OR UNKNOWN.THE COMPARISON SHOWS THE KNOWLEDGE OF TRAFFIC DIRECTION ONLY HELPS A LITTLE BIT IN DECREASING THE SUM OF CONNECTION DURATIONS.

2004

Cited by 23

### Table 1. The statistics distribution of 1-4 character words The table above shows that the number of three-character words adds up to 38,585, accounted for 44% of the total. The segmented words here are a little bit different from the lexical entries. But contrary to our expectation, the number of the unrecorded

2005

### Table 1: Detailed timings for vectorized monomial operations: Singular compiled with gcc version 2.7.2 for a Pentium Pro 200 (32 bit, little{endian) running Linux and average of timings for a HP C160 (32 bit, big{endian) running HP{UX 10.20 and a DEC Alpha (64 bit, little endian) running Linux.

1998

"... In PAGE 6: ... However, we shall not describe here the details of these routines, since their implementation is more or less obvious and they have less of an impact on the running time than the basic monomial operations. So much for the theory, now let us look at some ac- tual timings: Table1 shows various timings illustrating the e ects of the vectorized monomial operations described in this section. In the rst column, we list the used examples | details about those can be found in the Appendix.... ..."

Cited by 4

### Table X shows the results for the Minimal clearance heuris- tic. We set clmin to 0.5. We want to remark that we do not know the optimal values for these paths. Compared to the original paths, the lengths are reasonably short and compared to the Clearance+Partial shortcut heuristic, the lengths are (of course) a bit longer. Compared to Table IV, the clearance decreased a little bit.

2004

Cited by 17

### Table IX shows that indeed the total lengths are shorter when the clearance is increased in advance. Only the length of the Simple corridor scene was a little bit deteriorated, which is probably caused by the fact that the path was already close to the optimal path. The table also shows that it is harder to remove rotational motions than translational ones. We will study this further in future work.

2004

Cited by 17