Results 1 - 10 of 49,081

Table 1. Bottom-up version of the binary algorithm

in Enhancing simple power-analysis attacks on elliptic curve cryptosystems
by Elisabeth Oswald 2002
Cited by 9

Table 3.7: Bottom-up algorithm, extraction

in unknown title
by unknown authors

Table 3.8: Bottom-up algorithm, assembly

in unknown title
by unknown authors

Table 3 Levels and corresponding number of candidates employed by HD and BOTTOM-UP in Example 12.

in Hierarchical model-based diagnosis based on structural abstraction
by Luca Chittaro, Roberto Ranon 2004
"... In PAGE 36: ... Then, diagnosing sc1 with the remaining modes does not allow to remove any of them. The levels used by HD and by the BOTTOM-UP extension with the associated number of components that have to be considered are shown in in Table3 . The two algorithms exploit the same levels, but BOTTOM-UP is able to reduce the number of candidates that have to be considered at each level.... ..."
Cited by 3

Table 1 shows the algorithm for bottom-up evaluation. The main loop iterates over the set a26 of ground predicates.

in Scalable knowledge acquisition through memory organization
by David J. Stracuzzi 2005
"... In PAGE 5: ...Set a26 of ground predicates Example a18a20a3 a8 a6a5a8a7a22a21a23a7a22a24 a11 Algorithm: BottomUpa8 a26 a7 a18 a11 //main loop a25 a1a0 objects in a24 for each a1 a30 a26 do a2 a0 set of all bindings between a25 and a9a14a11 for each a24 a28a31a30 a2 do BUEvaluatora8 a1 a7a22a24 a28 a11 BUEvaluatora8 a1 a7a22a24 a11 //helper procedure a2 a0 set of bindings between a25 and a9a14a11 given a24 for each a24 a28a31a30 a2 do bind a24 a28 to a9a17a11 and evaluate a1 for each a1a66a28a31a30 a15a16a11 do if a1 a28 is learned and a1 improves a1 a28 then a24 a28 a28 a3a0a5a4a7a6a9a8 a10a12a11a14a13a16a15a17a11a14a13a12a18a20a19 a8 a1 a7 a1a66a28 a11 BUEvaluatora8 a1a29a28 a7a22a24 a28 a28 a11 Table1 . Online algorithm for bottom-up evaluation.... ..."
Cited by 2

Table 6: The SWAB (Sliding Window and Bottom-up) algorithm Using the buffer allows us to gain a semi-global view of the data set for Bottom-Up. However, it important to impose upper and lower bounds on the size of the window. A buffer that is allowed to grow arbitrarily large will revert our algorithm to pure Bottom-Up, but a small

in Segmenting Time Series: A Survey and Novel Approach
by Eamonn Keogh, Selina Chu, David Hart, Michael Pazzani 2003
Cited by 16

Table 1 Comparison of the runtimes of the bi-directional and bottom-up fd-algorithms for various datasets, and the impact of the noise threshold on the size of covers and on the computation time of algorithms.

in Database Dependency Discovery: A Machine Learning Approach
by Peter A. Flach, Iztok Savnik
"... In PAGE 18: ... For this reason, the algorithm may be of limited practical use for the discovery of dependen- cies from very large relations (depending on the num- ber and the type of attributes, as well). Table1 shows that in most cases, the time for com- puting the negative cover increases with the increase of the noise threshold. One of the reasons for this is the increased number of invalid dependencies in the negative cover which appears because of the relaxed condition for the invalid dependencies.... ..."

Table 3.3: A bottom-up partitioning algorithm. The algorithm takes as input the sample S, an attribute A, a set of thresholds T , and outputs a partition of S.

in Efficient Range Partitioning in Classification Learning
by Juho Rousu 2001
Cited by 4

Table 2. MUX Decomposition Results for a Simulated 32-to-1 MUX in a MIPS Processor BOTTOM-UP0 TOP-DOWN0 HYBRID0

in IMPROVED ALGORITHMS FOR LOW POWER MULTIPLEXOR DECOMPOSITION ∗
by Shaofa Yang, Hon Wai Leong, Jaffar Joxan, Shaofa Yang, Hon Wai Leong 2006
"... In PAGE 13: ... We then ran this simulated MUX design through the various MUX decompo- sition algorithms, followed by the post-optimization procedures. The results ob- tained are shown in Table2 . Table 2 shows rstly, that all the MUX decomposition algorithms are e ective in reducing the power dissipations over the default decom- position.... In PAGE 13: ... Table 2 shows rstly, that all the MUX decomposition algorithms are e ective in reducing the power dissipations over the default decom- position. Secondly, the general trends in Table2 and Table 3 are very similar. This suggests that the performance of our algorithms in real MUXes will be similar to cThe values of for BOTTOM-UP0 are calculated based on optimal non-uniform decompositions.... ..."

Table 3. Power Reduction obtained by LEVEL-POST and GREEDY-POST. BOTTOM-UP0 TOP-DOWN0 HYBRID0

in IMPROVED ALGORITHMS FOR LOW POWER MULTIPLEXOR DECOMPOSITION ∗
by Shaofa Yang, Hon Wai Leong, Jaffar Joxan, Shaofa Yang, Hon Wai Leong 2006
"... In PAGE 13: ...Table3 (pages 14{18) shows the power reduction obtained. Each entry in the ta- ble shows the power dissipation in the form of , the percentage above the optimal power dissipation.... In PAGE 13: ... Table 2 shows rstly, that all the MUX decomposition algorithms are e ective in reducing the power dissipations over the default decom- position. Secondly, the general trends in Table 2 and Table3 are very similar. This suggests that the performance of our algorithms in real MUXes will be similar to cThe values of for BOTTOM-UP0 are calculated based on optimal non-uniform decompositions.... In PAGE 17: ...Table3 . (Continued).... In PAGE 18: ...Table3 . (Continued).... In PAGE 19: ...Table3 . (Continued).... In PAGE 20: ...Table3 . (Continued).... ..."
Results 1 - 10 of 49,081