Compiling Constraint Networks into Multivalued Decomposable Decision Graphs

Authors: Frédéric Koriche, Jean-Marie Lagniez, Pierre Marquis, Samuel Thomas

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Intensive experiments showed that our compiler cn2mddg succeeds in compiling CNs which are out of the reach of standard approaches based on a translation of the input network to CNF, followed by a compilation to Decision-DNNF.
Researcher Affiliation Academia Fr ed eric Koriche, Jean-Marie Lagniez, Pierre Marquis, Samuel Thomas CRIL-CNRS and Universit e d Artois, Lens, France {koriche, lagniez, marquis, thomas}@cril.univ-artois.fr
Pseudocode Yes Algorithm 1 provides the pseudo-code for the compiler cn2mddg.
Open Source Code Yes The run-time code of our compiler, as well as the translators and the benchmarks used in our experiments, and additional empirical results, can be downloaded from www.cril.fr/KC/
Open Datasets Yes We have considered 173 CNs from 15 data sets, downloaded from github.com/Mini Zinc/minizincbenchmarks, www.cril.univ-artois.fr/ lecoutre/benchmarks.html, and www.itu.dk/research/cla/externals/clib/.
Dataset Splits No The paper uses
Hardware Specification Yes Our experiments have been conducted on a Quadcore Intel XEON X5550 with 32GB of memory.
Software Dependencies No The paper mentions software components and formats like 'XCSP 2.1 format', 'c2d', 'Dsharp', 'Sugar', and 'Azucar', but it does not provide specific version numbers for these or other software dependencies required to reproduce the experiments.
Experiment Setup No The paper describes the techniques and heuristics used in their compiler (e.g., caching, variable ordering based on betweenness centrality, universal constraint handling) but does not provide specific hyperparameter values or concrete system-level training/experimental settings typical for reproducibility (e.g., learning rates, batch sizes, epochs).