Fully Online Matching with Stochastic Arrivals and Departures

Authors: Zihao Li, Hao Wang, Zhenzhen Yan

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive numerical studies to evaluate the performance of our algorithms.
Researcher Affiliation Academia Zihao Li, Hao Wang, Zhenzhen Yan Nanyang Technological University zihao004@e.ntu.edu.sg, hao wang@ntu.edu.sg, yanzz@ntu.edu.sg
Pseudocode Yes Algorithm 1: SAM(γ) Input: Online arrivals of agents Parameter: Scaling parameter γ (0, 1]
Open Source Code No The paper does not provide a direct link to open-source code for the described methodology or state that the code will be made publicly available. It only mentions using Gurobi as a solver.
Open Datasets No The paper describes generating synthetic datasets for experiments but does not provide access information (link, citation, or repository) for these generated datasets to be publicly available.
Dataset Splits No The paper discusses evaluating performance over "realized sequences" and states "For each parameter setting, we test |R| = 50 realized sequences." However, it does not specify traditional train/validation/test dataset splits, which are not directly applicable to its online problem setup.
Hardware Specification Yes We use a computer with 2.2 GHz Intel Core i7 processor, 16 GB 1600 MHz DDR3 memory and Intel Iris Pro 1536 MB Graphics to run all the experiments.
Software Dependencies No The paper mentions "We use Gurobi (Gurobi Optimization 2022) as our solver." While Gurobi is a specific software, it provides the year of the reference (2022) but not a numerical version number (e.g., Gurobi 9.5).
Experiment Setup Yes The parameters are q = 2.5 and geometric distribution with P G = 0.5. We set T = 3000 which is much larger than the sojourn time of any vertex under any tested distribution.