An Experimentally Efficient Method for (MSS,CoMSS) Partitioning

Authors: Eric Grégoire, Jean-Marie Lagniez, Bertrand Mazure

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive empirical evaluation shows that it is more robust and more efficient on most instances than currently available techniques. All experimentations have been conducted on Intel Xeon E5-2643 (3.30GHz) processors with 7.6Gb RAM on Linux Cent OS.
Researcher Affiliation Academia Eric Gr egoire, Jean-Marie Lagniez, Bertrand Mazure CRIL Universit e d Artois & CNRS rue Jean Souvraz F-62307 Lens, France {gregoire,lagniez,mazure}@cril.fr
Pseudocode Yes Algorithm 1: BLS (Basic Linear Search) and Algorithm 2: CMP (Computational Method for Partitioning) are provided.
Open Source Code Yes CMP and all experimentation data are available from www.cril.univ-artois.fr/documents/cmp/.
Open Datasets Yes The first one was made of the 1343 benchmarks used and referred to in (Marques-Silva et al. 2013): they are small-sized industrial-based instances from SAT competitions www.satcompetition.org and structured instances from the MAXSAT evaluations maxsat.ia.udl.cat:81. We enriched this experimentation setting by also considering a second series of benchmarks, made of all the 295 instances used for the 2011 MUS competition organized in parallel with the SAT one.
Dataset Splits No The paper mentions using established benchmarks but does not specify how the data was split into training, validation, or testing sets, nor does it refer to a standard split that includes a validation set for these benchmarks.
Hardware Specification Yes All experimentations have been conducted on Intel Xeon E5-2643 (3.30GHz) processors with 7.6Gb RAM on Linux Cent OS.
Software Dependencies No The paper mentions "MINISAT (E en and S orensson 2004)" as the CDCL SAT-solver, but it does not provide a specific version number for MINISAT itself, nor does it list any other software dependencies with version numbers.
Experiment Setup Yes Time limit was set to 30 minutes.