Algorithms for Trip-Vehicle Assignment in Ride-Sharing

Authors: Xiaohui Bei, Shengyu Zhang

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments are conducted showing that our algorithm actually has a much better approximation ratio (around 1.2) on synthetically generated data.
Researcher Affiliation Academia Xiaohui Bei School of Physical and Mathematical Sciences Nanyang Technological University Singapore Shengyu Zhang Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong
Pseudocode Yes Algorithm 1 Allocation(G, LR, LD) Input: A nonnegative weighted graph G = (V, E, w), request locations LR = {(si, ti) V 2 : i R}, driver locations LD = {dk V : d D}. Output: An allocation M = {(k, {i, j}) : k D, i, j R}.
Open Source Code No The paper does not provide any statement or link indicating that the source code for their methodology is publicly available.
Open Datasets No The paper uses synthetically generated data, not a publicly available dataset. "The inputs {si, ti, dk : i R, k D} in two ways. In the first, we generate all si, ti, dk independently from uniform distribution on C. The second distribution for inputs si, ti, dk is Gaussian mixture..." There is no link or citation to a public dataset or data generation tool.
Dataset Splits No The paper describes how the data for experiments is generated (uniform and Gaussian mixture distributions) but does not specify any training, validation, or test dataset splits in the context of model training or evaluation. The evaluation is done on the generated data against a lower bound.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, memory specifications, or cloud computing instance types used for running the experiments.
Software Dependencies No The paper does not mention any specific software dependencies or their version numbers (e.g., programming languages, libraries, or solvers).
Experiment Setup Yes The paper describes parameters used for generating experimental data, which are part of the experimental setup for a combinatorial optimization problem. "We consider region C def = [0, B] [0, B] for a big number B (such as 100). We tested two distance measures: ℓ1norm and ℓ2-norm... In the first, we generate all si, ti, dk independently from uniform distribution on C. The second distribution for inputs si, ti, dk is Gaussian mixture, where we have a number of centers {μ1, . . . , μc} and covariance matrices Σ1, . . . , Σc." It also specifies ranges for n (10-50), B (10-100), σ (1-10), and c (1-10 centers).