Optimal Aggregation of Uncertain Preferences

Authors: Ariel Procaccia, Nisarg Shah

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

Reproducibility Variable Result LLM Response
Research Type Experimental Our experimental results in Section 5 indicate that this is true even with preferences from real-world datasets.Specifically, we use five datasets from Preflib (Mattei and Walsh 2013): AGH Course Selection (D1), Netflix (D2), Skate (D3), Sushi (D4), and T-Shirt (D5). We use CPLEX to find the minimum feedback arc set through integer linear programming.
Researcher Affiliation Academia Ariel D. Procaccia Computer Science Department Carnegie Mellon University arielpro@cs.cmu.edu Nisarg Shah Computer Science Department Carnegie Mellon University nkshah@cs.cmu.edu
Pseudocode No The paper does not include any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement or link for open-source code for the methodology described.
Open Datasets Yes Specifically, we use five datasets from Preflib (Mattei and Walsh 2013): AGH Course Selection (D1), Netflix (D2), Skate (D3), Sushi (D4), and T-Shirt (D5).
Dataset Splits No The paper describes a simulation setup where uncertainty is introduced into existing profiles and an approximation ratio is computed over 1000 simulations, but it does not specify traditional training/validation/test dataset splits.
Hardware Specification No The paper mentions using 'CPLEX' (software) but does not provide any specific details about the hardware used for the experiments (e.g., GPU/CPU models, memory).
Software Dependencies No We use CPLEX to find the minimum feedback arc set through integer linear programming. (No version number for CPLEX is provided).
Experiment Setup Yes For each preference profile, we compute the approximation ratio averaged over 1000 simulations. In each simulation each vote in the profile is converted into an uncertain vote, represented as the Mallows model whose central ranking is the vote itself and whose noise parameter ϕ is chosen uniformly at random from [0, 1].