Steiner Tree Problems with Side Constraints Using Constraint Programming

Authors: Diego de Uña, Graeme Gange, Peter Schachte, Peter Stuckey

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Section 6 shows the experimental results solving the pure STP and two of its variants (comparing different versions of our propagator against the CHOCO3 solver).
Researcher Affiliation Academia 1 Department of Computing and Information Systems The University of Melbourne 2 National ICT Australia, Victoria Laboratory {d.deunagomez@student.,gkgange@,schachte@,pstuckey@}unimelb.edu.au
Pseudocode No The paper describes algorithms (e.g., 'reachable', 'articulations') but does not present them in formal pseudocode or algorithm blocks.
Open Source Code No The paper does not provide explicit statements or links for the open-source code of their methodology.
Open Datasets Yes The benchmarks used in this study are from the Stein Lib (Koch, Martin, and Voß 2000).
Dataset Splits No The paper uses benchmarks from Stein Lib (e.g., ES10FST, ES20FST, B) but does not specify how these datasets are split into training, validation, and test sets.
Hardware Specification Yes All tests were run on a Linux 3.16 Intel R Core TM i7-4770 CPU @ 3.40GHz, 15.6GB of RAM machine.
Software Dependencies Yes We modelled the pure STP and two variations in MINIZINC and solved them with the CHUFFED solver. We used the latest CHOCO3 (Prud homme, Fages, and Lorca 2014) solver as a comparison...we solve it using CPLEX 12.4.
Experiment Setup Yes We used the same search strategy in all the implementations. The order of the variables is: edges sorted by decreasing weight, then nodes in arbitrary order. The value strategy is: try assigning the values {false, true} in that order to each variable. This is the strategy that gave the best results.