Subset Approximation of Pareto Regions with Bi-objective A*

Authors: Nicolás Rivera, Jorge A. Baier, Carlos Hernández10345-10352

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, we use standard roadmap benchmarks to show that, depending on the chosen parameters, BOA* can compute about 10–20% of the solutions in about one order of magnitude less time.Our experimental evaluation had the objective of evaluating the performance of BOA* run over Pα,β using different (α, β) values over a large number of instances in several standard road maps.
Researcher Affiliation Academia 1Instituto de Ingenier ıa Matem atica, Universidad de Valpara ıso, Valpara ıso, Chile 2Departamento de Ciencia de la Computaci on, Pontificia Universidad Cat olica de Chile, Santiago, Chile 3Instituto Milenio Fundamentos de los Datos, Santiago, Chile 4Facultad de Ingenier ıa y Tecnolog ıa, Universidad San Sebasti an, Bellavista 7, 84205254, Santiago, Chile
Pseudocode Yes Algorithm 1: Bi-Objective A* (BOA*)
Open Source Code No The paper does not provide concrete access to source code (specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described in this paper.
Open Datasets Yes We evaluated our approach, implemented in C, on maps of the 9th DIMACS Implementation Challenge: Shortest Path1; specifically, 50 random instances for each of four USA road maps used by Machuca and Mandow (2012).1http://users.diag.uniroma1.it/challenge9/download.shtml
Dataset Splits No The paper evaluates its approach on '50 random instances' from a benchmark but does not specify training, validation, or test dataset splits needed for reproducibility.
Hardware Specification Yes We ran all experiments on a 3.80GHz Intel(R) Core(TM) i7-10700K CPU Linux machine with 64GB of RAM.
Software Dependencies No The paper states 'implemented in C' but does not provide specific version numbers for any compilers, libraries, or other software dependencies needed to replicate the experiment.
Experiment Setup Yes Our experimental evaluation had the objective of evaluating the performance of BOA* run over Pα,β using different (α, β) values over a large number of instances in several standard road maps. We evaluated our approach, implemented in C, on maps of the 9th DIMACS Implementation Challenge: Shortest Path1; specifically, 50 random instances for each of four USA road maps used by Machuca and Mandow (2012).In each run we consider the same value for α and β.