A Robust Exact Algorithm for the Euclidean Bipartite Matching Problem

Authors: Akshaykumar Gattani, Sharath Raghvendra, Pouyan Shirzadian

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we present the results of our experiments comparing the execution time of our algorithm to that of the Hungarian algorithm.
Researcher Affiliation Academia Akshaykumar G. Gattani1, Sharath Raghvendra1, and Pouyan Shirzadian1 1Department of Computer Science, Virginia Tech
Pseudocode Yes The pseudo-code of our divide-and-conquer algorithm is provided in Algorithm 1.
Open Source Code Yes Our implementations are available at https://github.com/agattani190/Exact-Euclidean-Bipartite-Matching.
Open Datasets Yes For a real-world dataset, we employ the New York Taxi dataset [47] and obtain two distributions, namely (i) the distribution of pickup locations (Pickup) and (ii) the distribution of drop-off locations (Drop-off) of passengers. We filtered the datasets by considering trips in seven dates in 2014 with (i) a trip duration of at least 3 minutes, and (ii) a trip velocity of at most 110mph. ... [47] NYC Taxi and Limousine Commission (TLC). Trip record data. https://www.nyc.gov/ site/tlc/about/tlc-trip-record-data.page, 2023. Accessed: 2023-03-01.
Dataset Splits No The paper describes the synthetic and real-world datasets used and how samples were drawn but does not specify train, validation, or test splits by percentages or sample counts, nor does it mention cross-validation.
Hardware Specification Yes All computations are performed using a single calculation thread on a computer with a 2.6 GHz 6-Core Intel Core i7 CPU and 16 GB of 2667 MHz DDR4 RAM.
Software Dependencies No The paper states "Both algorithms are implemented in Java" and "we use the classical implementation of Dijkstra s shortest path algorithm" but does not provide specific version numbers for Java or any other software components.
Experiment Setup Yes For the synthetic dataset, we use two distributions, namely (i) a uniform distribution defined on the unit square (Uniform), and (ii) a Gaussian distribution constrained to the unit square with a randomly chosen mean inside the unit square and a standard deviation of 0.25 (Gaussian). For a real-world dataset, we employ the New York Taxi dataset [47] and obtain two distributions... We filtered the datasets by considering trips in seven dates in 2014 with (i) a trip duration of at least 3 minutes, and (ii) a trip velocity of at most 110mph. ... In each test, we conducted experiments using two sets of n i.i.d samples from distributions µ and ν within the unit square.