Online Spatio-Temporal Matching in Stochastic and Dynamic Domains

Authors: Meghna Lowalekar, Pradeep Varakantham, Patrick Jaillet

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

Reproducibility Variable Result LLM Response
Research Type Experimental Traditionally, greedy myopic approaches have been adopted to address such large scale online matching problems. While they provide solutions in a scalable manner, due to their myopic nature the quality of matching obtained can be improved significantly (demonstrated in our experimental results). In this paper, we present a two stage stochastic optimization formulation to consider expected future demand. We then provide multiple enhancements to solve large scale problems more effectively and efficiently. Finally, we demonstrate the significant improvement provided by our techniques over myopic approaches on two real world taxi data sets.
Researcher Affiliation Academia School of Information Systems, Singapore Management University Dept. of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, USA
Pseudocode No The paper includes mathematical formulations and descriptions of algorithms but does not provide structured pseudocode or algorithm blocks labeled as such.
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper states: 'We conducted our experiments by taking demand distribution from real world datasets (henceforth referred to as Dataset1 and Dataset2) of two taxi companies6 in a big asian city.' However, no concrete access information (link, DOI, specific citation for public access) is provided for these datasets.
Dataset Splits No The paper describes its evaluation approach: 'Once the assignment is computed, we evaluate the assignment on realized requests (which are samples from past data that are not considered while computing the assignment at t + 1).' This implies a form of splitting, but it does not specify exact percentages, sample counts, or refer to predefined validation splits, which are required for reproducibility.
Hardware Specification No The paper states: 'Moreover, we performed our experiments on academic systems,on commercial systems, runtime should be much lower.' This is a general statement, but it does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts.
Software Dependencies No The paper does not provide specific software names with version numbers or library dependencies used in the experiments.
Experiment Setup Yes To provide the right tradeoff between run-time and solution quality, we terminated BD after 3 iterations.