Expected Frequency Matrices of Elections: Computation, Geometry, and Preference Learning

Authors: Niclas Boehmer, Robert Bredereck, Edith Elkind, Piotr Faliszewski, Stanisław Szufa

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We use the map of elections approach of Szufa et al. (AAMAS-2020) to analyze several well-known vote distributions. For each of them, we give an explicit formula or an efficient algorithm for computing its frequency matrix, which captures the probability that a given candidate appears in a given position in a sampled vote. We use these matrices to draw the skeleton map of distributions, evaluate its robustness, and analyze its properties. Finally, we develop a general and unified framework for learning the distribution of real-world preferences using the frequency matrices of established vote distributions.
Researcher Affiliation Academia Niclas Boehmer Algorithmics and Computational Complexity Technische Universität Berlin niclas.boehmer@tu-berlin.de Robert Bredereck TU Clausthal robert.bredereck@tu-clausthal.de Edith Elkind University of Oxford elkind@cs.ox.ac.uk Piotr Faliszewski AGH University faliszew@agh.edu.pl Stanisław Szufa AGH University Jagiellonian University szufa@agh.edu.pl
Pseudocode No No pseudocode or algorithm blocks were found.
Open Source Code Yes The source code used for the experiments is available in a Git Hub repository1. 1github.com/Project-PRAGMA/Expected-Frequency-Matrices-Neur IPS-2022
Open Datasets Yes We consider elections from the real-world datasets used by Boehmer et al. (2021b). They generated 15 elections with 10 candidates and 100 voters (with strict preferences) from each of the eleven different real-world election datasets (so, altogether, they generated 165 elections, most of them from Preflib (Mattei & Walsh, 2013)).
Dataset Splits No The paper does not provide specific details on training, validation, or test splits for the datasets used.
Hardware Specification No No specific hardware details (e.g., CPU, GPU models, or cloud resources) were mentioned for running experiments.
Software Dependencies No The paper mentions 'Python sklearn.manifold.MDS package' but does not provide specific version numbers for software dependencies.
Experiment Setup Yes Let Φ = {0, 0.05, 0.1, . . . , 1} be a set of normalized dispersion parameters that we will be using for Mallows-based distributions in this section.