On Robustness in Qualitative Constraint Networks

Authors: Michael Sioutis, Zhiguo Long, Tomi Janhunen

IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we report on a preliminary experimentation that was performed primarily to assess the differences that may or may not exist between the scenarios of a given QCN N with respect to their similarity measure and perturbation tolerance. Secondarily, results are reported on the time needed to compute a robust scenario of N, on the size of [[N]], and on the % of the time that a maximum scenario of N is satisfiable and hence also a robust scenario of N.
Researcher Affiliation Academia 1Otto-Friedrich-University Bamberg, WIAI, Bamberg, Germany 2Southwest Jiaotong University, SIST & IAI, Chengdu, China 3Tampere University, ICT, Tampere, Finland
Pseudocode Yes Algorithm 1: Robust Scen(N, Oracle)
Open Source Code No The paper does not include an unambiguous statement that the authors are releasing the code for the described methodology, nor does it provide a direct link to a source-code repository.
Open Datasets Yes We considered 100 satisfiable QCNs of 50 constraints each that were created using uniformly selected interval relations appearing in job-shop scheduling problems in the SMT-LIB [Barrett et al., 2016];
Dataset Splits No The paper describes the datasets used (QCNs from SMT-LIB and standard interval relations) but does not provide specific details on how these were split into training, validation, or test sets, or specify cross-validation settings.
Hardware Specification Yes We used a computer with an Intel R Xeon R CPU E3-1231 v3 processor at 3.40GHz per core, 16 GB of RAM, and the Xenial Xerus x86 64 OS (Ubuntu Linux).
Software Dependencies Yes All algorithms were coded in Python and run using Py Py 7.1.1.
Experiment Setup No The paper does not provide specific hyperparameters (e.g., learning rate, batch size, epochs), model initialization, or training schedules. It only describes the general computational environment and datasets.