• Documents
  • Authors
  • Tables
  • Log in
  • Sign up
  • MetaCart
  • DMCA
  • Donate

CiteSeerX logo

Tools

Sorted by:
Try your query at:
Semantic Scholar Scholar Academic
Google Bing DBLP
Results 1 - 10 of 48,586
Next 10 →

Table 3. Top-down evidences for gutters

in Integrating Models for Object Reconstruction
by Wolfram Willuhn
"... In PAGE 32: ... We therefore wanted the uncertainty information to freely flow between all models but due to stability considerations restricted this to two unidirectional bottom-up and top-down flows. In order to decide which of the contradicting gutters (G1 or G2) is the correct one, we look at the top-down evi- dence ( Table3 ). G1 has the advantage that is better fits with the two roof gables than the shorter G2.... ..."

Table 3: The cost of labeling with various Cable strategies: Optimal (Opt), Expert (Exp), Top-down (Top-down), Bottom-up (Btm- up), and Random (Rand). These costs are compared to the Baseline method (Base), which does not use Cable.

in Debugging Temporal Specifications with Concept Analysis
by Glenn Ammons, David Mandelin, Rastislav Bodík, James R. Larus 2003
"... In PAGE 9: ... 5.3 Traversal strategies Table3 compares the cost of labeling by a variety of meth- ods, where cost is defined as in Section 4.2.... In PAGE 9: ... Then, we measured the cost of obtaining the same labeling with each method. Because the Top-down, Bottom-up, and Random strate- gies have non-deterministic costs, Table3 reports the lowest cost for Bottom-up and the arithmetic mean and standard deviation of the cost of 1024 trials for Top-down and Random. We were un- able to measure the cost of the Optimal strategy for RegionsBig and XSaveContext, because the program we wrote to evaluate the strategies on these specifications took too long to run.... In PAGE 9: ... In addition to the strategies listed in Section 4.2, Table3 lists two other methods: Expert This method measured the actual cost of labeling for the expert user. The expert used a mostly top-down approach, but sometimes directed his search based on transitions he found interesting.... In PAGE 12: ... However, Cable was often significantly better in terms of labeling cost, which reflects the number of concepts that the expert examined. Comparison with Table3 reveals three cases where one expert beat the other by a large margin: on XtFree, the expert who used the traversal strategy beat the expert who used navigation (28 to 42); on XGContextFromGC and XSetFont, the expert who used navigation won (12 to 24 and 12 to 53). On XGContextFromGC, the navigation expert even beat Optimal, which was possible be- cause the navigation expert was working with a larger concept lattice.... ..."
Cited by 35

Table IV. Top-Down Incremental Deletion

in Incremental Analysis of Constraint Logic Programs
by Manuel Hermenegildo, German Puebla, Kim Marriott, Peter J. Stuckey 1996
Cited by 74

Table 3: The generic Top-Down algorithm

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

Table 1: topDown policy conditions

in Generalized xml security views
by Gabriel Kuper, Gabriel Kuper, Fabio Massacci, Fabio Massacci, Nataliya Rassadko, Nataliya Rassadko 2005
Cited by 5

Table 1. The top-down induction algorithm for PCTs.

in Ensembles of Multi-Objective Decision Trees
by Dragi Kocev, Celine Vens, Jan Struyf
"... In PAGE 3: ... PCTs can be constructed with a standard top-down induction of decision trees (TDIDT) algorithm [19]. The algorithm is shown in Table1 . The heuristic that is used for selecting the tests is the reduction in variance caused by parti- tioning the instances (see line 4 of BestTest).... In PAGE 4: ...1 Ensembles for Multi-Objective Decision Trees In order to apply bagging to MODTs, the procedure PCT(Ei) (Table 1) is used as a base classifier. For applying random forests, the same approach is followed, changing the procedure BestTest ( Table1 , right) to take a random subset of size f(x) of all possible attributes. In order to combine the predictions output by the base classifiers, we take the average for regression, and apply a probability distribution vote instead of a simple majority vote for classification, as suggested by Bauer and Kohavi [23].... ..."

Table 1. Comparison of SOA development methodologies. (A relative quantitative scale 1-5 is used for some criteria. Also, M = Meet-in-the-Middle, T = Top-Down, B = Bottom-Up, and ? = No Data)

in Preface
by Stephen Gorton, Monika Solanki, Stephen Reiff-marganiec 2007
"... In PAGE 27: ... 3. Parallel We summarize the translation in the Table1 showing the relation between language constructs. Table 1.... In PAGE 27: ...ig. 3. Parallel We summarize the translation in the Table 1 showing the relation between language constructs. Table1 . Synoptic table of the translation WS-CDL PROMELA participantType a PROMELA process channelVariable one channel or two channels if the channelVariable is used for passing channels activity statement in the form: channelvariable[!|?]exchange name, informationType sequence sequence of statements choice if statement inside the involved processes repetitions (formally workunit) do statement inside the involved processes parallel... In PAGE 82: ...Annapaola Marconi et al. a b forwarder: simply forwards data received on the input node to the output node f c a b function: upon receiving data on all input nodes, computes the function result and forwards it to the output node a c b fork: forwards data received on the input node to all the output nodes c a b merge: forwards data received on some input node to the output node, preserving temporal order a b + cloner: forwards, one or more times, data received from the input node to the output node a b ? lter: receives data on the input node and either forwards it to the output node or discards it a c b X xor: forwards data received on the input node to (exactly) one of the output nodes Table1 . Basic elements of the data- ow requirements speci cation language.... ..."

Table 2: Top-down Hierarchical topology Parameters.

in BRITE: Universal Topology Generation from a User's Perspective
by Alberto Medina, Anukool Lakhina, Ibrahim Matta, John Byers 2001
Cited by 78

Table 2.11: Description of Top-down Parsing

in Chapter 2 Syntax
by unknown authors
"... In PAGE 46: ...rammar.29 From the viewpoint of the human reader, the definitions are clutter. From the viewpoint of implementation, the definitions are treated as suggested for L, D and C in Gem: predefined and to be implemented as efficiently as the hardware will allow30. 29EoL stands for end-of-line; EoF stands for end-of-file 30See Table2... ..."

Table 2.12: A Top-down Proof

in Chapter 2 Syntax
by unknown authors
"... In PAGE 46: ...rammar.29 From the viewpoint of the human reader, the definitions are clutter. From the viewpoint of implementation, the definitions are treated as suggested for L, D and C in Gem: predefined and to be implemented as efficiently as the hardware will allow30. 29EoL stands for end-of-line; EoF stands for end-of-file 30See Table2... ..."
Next 10 →
Results 1 - 10 of 48,586
Powered by: Apache Solr
  • About CiteSeerX
  • Submit and Index Documents
  • Privacy Policy
  • Help
  • Data
  • Source
  • Contact Us

Developed at and hosted by The College of Information Sciences and Technology

© 2007-2019 The Pennsylvania State University