Eliciting Kemeny Rankings

Authors: Anne-Marie George, Christos Dimitrakakis

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental All described methods are compared on synthetic data. Finally, we analyse the sample complexity experimentally on synthetic data for all developed methods.
Researcher Affiliation Academia 1University of Oslo, Norway 2University of Neuchatel, Switzerland
Pseudocode Yes Algorithm 1: Kemeny El(Q, ρ, δ): Sampling with Replacement from Q Q(k) Algorithm 2: Kemeny El(Q, ρ, δ): Sampling w/o Replacement from Q P(k, n) Algorithm 3: Pruning CI( ˆQ, C, C)
Open Source Code Yes All code is written in Python and is publicly available under https://github.com/annemage/Eliciting-Kemeny-Rankings.
Open Datasets No The paper uses synthetically generated data. "We generate preference matrices Q P(k, n) for n = 10 voters uniformly at random for up to k = 9 arms." No concrete access information for a publicly available dataset is provided.
Dataset Splits No The paper uses synthetically generated data. It does not mention specific training, validation, or test dataset splits in terms of percentages or sample counts. The experiment focuses on sample complexity for approximation bounds.
Hardware Specification No No specific hardware details such as GPU models, CPU types, or memory amounts used for running experiments are mentioned in the paper.
Software Dependencies No The paper states "All code is written in Python" but does not provide specific version numbers for Python or any other software libraries or dependencies used in the experiments.
Experiment Setup Yes Setup: We generate preference matrices Q P(k, n) for n = 10 voters uniformly at random for up to k = 9 arms. We set an approximation value of ρ = 0.1 k(k 1) / 2 , i.e., 10% of the worst case difference in approximated and optimal Kemeny score, and approximation probability (1 δ) = 0.95.