Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].

Revisiting Counting Solutions for the Global Cardinality Constraint

Authors: Giovanni Lo Bianco, Xavier Lorca, Charlotte Truchet, Gilles Pesant

JAIR 2019 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Section 5: Experimental Analysis : E๏ฌƒciency of the New Upper Bound within Counting-Based Search. We generated several random CSP, with the method described above with the following parameters : n = 20, m = 20, an edge density p {0.33, 0.66, 1.0}, ฯ„ [0.1, 1] by steps of 0.1 and n C [2, 10] by steps of 1. For each couple (ฯ„, n C), we generate 50 random instances for a total of 4500 instances.
Researcher Affiliation Academia Giovanni Lo Bianco EMAIL IMT Atlantique, 4 Rue Alfred Kastler, 44300 Nantes, France; Xavier Lorca EMAIL IMT Mines Albi, All ee des Sciences, 81000 Albi, France; Charlotte Truchet EMAIL UFR de Sciences et Techniques, 2, rue de la Houssini ere, BP 92208, 44322 NANTES CEDEX 3, France; Gilles Pesant EMAIL Polytechnique Montr eal, 2900 Boulevard Edouard-Montpetit, Montr eal, QC H3T 1J4, Canada. All listed affiliations correspond to academic institutions.
Pseudocode Yes Algorithm 1 Compute UBIP(X, ฯ‰)
Open Source Code No The paper does not provide any explicit statement about releasing its own source code, nor does it provide a link to a code repository for the methodology described. It only mentions using a third-party solver: "The instances and the strategies are implemented in Choco solver (Prud homme, Fages, & Lorca, 2017)".
Open Datasets No The paper describes generating its own random instances for the experiments: "We generated several random CSP, with the method described above with the following parameters : n = 20, m = 20, an edge density p {0.33, 0.66, 1.0}, ฯ„ [0.1, 1] by steps of 0.1 and n C [2, 10] by steps of 1. For each couple (ฯ„, n C), we generate 50 random instances for a total of 4500 instances." There is no indication that these generated instances are made publicly available or that existing public datasets were used.
Dataset Splits No The paper describes generating random instances for experiments but does not mention any training/test/validation splits for these instances or any other dataset. The text states: "For each couple (ฯ„, n C), we generate 50 random instances for a total of 4500 instances."
Hardware Specification Yes The instances and the strategies are implemented in Choco solver (Prud homme, Fages, & Lorca, 2017) and we set, for each resolution, the time limit to 1 min and run on a 2.2GHz Intel Core i7 with 2.048GB.
Software Dependencies No The paper mentions the use of "Choco solver (Prud homme, Fages, & Lorca, 2017)", but this citation points to "Choco Documentation" and does not provide a specific version number for the Choco solver itself. No other software dependencies with version numbers are listed.
Experiment Setup Yes We generated several random CSP, with the method described above with the following parameters : n = 20, m = 20, an edge density p {0.33, 0.66, 1.0}, ฯ„ [0.1, 1] by steps of 0.1 and n C [2, 10] by steps of 1. For each couple (ฯ„, n C), we generate 50 random instances for a total of 4500 instances. We solve each instance with two different strategies: max SD with the PQZ bound and max SD with the corrected bound. ... and we set, for each resolution, the time limit to 1 min.