Evaluation of Project Performance in Participatory Budgeting

Authors: Niclas Boehmer, Piotr Faliszewski, Łukasz Janeczko, Dominik Peters, Grzegorz Pierczyński, Šimon Schierreich, Piotr Skowron, Stanisław Szufa

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

Reproducibility Variable Result LLM Response
Research Type Experimental Second, we compute our measures for projects from real-life PB instances from Pabulib [Faliszewski et al., 2023] and analyze their behavior. As a side result, we obtain some striking examples of nonmonotonicity of proportionality-oriented PB rules. We also demonstrate the usefulness of our measures by giving a detailed analysis of the Wieliczka election.
Researcher Affiliation Academia Niclas Boehmer1 , Piotr Faliszewski2 , Łukasz Janeczko2 , Dominik Peters3 , Grzegorz Pierczy nski2,4 , ˇSimon Schierreich5 , Piotr Skowron4 and Stanisław Szufa2,3 1Harvard University 2AGH University, Krak ow 3CNRS, LAMSADE, Universit e Paris Dauphine PSL 4University of Warsaw 5Czech Technical University in Prague
Pseudocode No The paper describes algorithmic procedures in text and provides a 'Proof sketch (PHRAGMÉN)' section outlining steps for an algorithm, but it does not include a clearly labeled pseudocode or algorithm block/figure.
Open Source Code Yes The code for our experiments is available at github.com/Project-PRAGMA/ Project-Performance-IJCAI-2024.
Open Datasets Yes We conduct our experiments on all 551 PB instances with approval votes from Pabulib [Faliszewski et al., 2023]
Dataset Splits No The paper analyzes real-world participatory budgeting instances but does not describe any specific training, validation, or test dataset splits or their sizes, as it does not involve training a machine learning model in the conventional sense.
Hardware Specification Yes We ran our experiments on 10 threads of an Intel(R) Xeon(R) Gold 6338 CPU @ 2.00GHz core.
Software Dependencies No The paper mentions using 'Gurobi' for experiments but does not provide a specific version number for this or any other software dependency.
Experiment Setup Yes To compute the measures, we use the algorithms described in Section 3. For the adding-voters and sampling algorithms, we increase the approval score of the designated project by 1% in each step (repeating each step 100 times for the sampling algorithms).