Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
New Approximations for Coalitional Manipulation in Scoring Rules
Authors: Orgad Keller, Avinatan Hassidim, Noam Hazon
JAIR 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | While we believe that the main message of this work is the research of Rα-UCM and Rα-WCM from a theoretical perspective, i.e., establishing positive results regarding the approximability of Rα-UCM and Rα-WCM, we also wanted to check how they compare with the known methods, with both being of a greedy type. ... To evaluate our algorithms for both UCM and WCM in a well-studied setting, we ran experiments for Borda-UCM and Borda-WCM on sets of values for n, k and m... |
| Researcher Affiliation | Academia | Orgad Keller EMAIL Department of Computer Science Bar-Ilan University, Israel Avinatan Hassidim EMAIL Department of Computer Science Bar-Ilan University, Israel Noam Hazon EMAIL Department of Computer Science Ariel University, Israel |
| Pseudocode | Yes | Algorithm 1: UCM Approximation algorithm. ... Algorithm 2: WCM Approximation algorithm. |
| Open Source Code | Yes | We implemented our algorithm for the case of Rα-UCM and Rα-WCM and uploaded the code to a public repository.6 The main subroutine in our implementation is tasked with solving the LP dual that we defined. ... 6 https://github.com/okeller/Borda Manipulation |
| Open Datasets | No | We performed two sets of experiments, where we drew the non-manipulator votes from either a uniform distribution, or from a P olya-Eggenberger urn model, similar to the experiments performed by Davies et al. (2014). ... To model what we see as a relatively realistic setting, we randomly chose weights according to the Zeta distribution (or equivalently, the unbounded Zipf law)... The paper generates data from specified distributions rather than using external publicly available datasets with access information. |
| Dataset Splits | No | We conducted 100 for the UCM experiments and 90 for the WCM experiments. ... We performed two sets of experiments, where we drew the non-manipulator votes from either a uniform distribution, or from a P olya-Eggenberger urn model... The paper describes running multiple experiments by drawing votes from distributions based on varying parameters (n, k, m) rather than providing specific train/test/validation splits for a fixed dataset. |
| Hardware Specification | No | The paper discusses the implementation and empirical analysis of algorithms but does not provide specific details on the hardware used for running these experiments. |
| Software Dependencies | Yes | we simulated this by using general LP-solving libraries (Andersen, Dahl, & Vandenberghe, 2016; Makhorin, 2017) ... Andersen, M. S., Dahl, J., & Vandenberghe, L. (2016). CVXOPT: A Python package for convex optimization, version 1.1.9. cvxopt.org. ... Makhorin, A. (2017). GLPK: GNU linear programming kit, version 4.61. www.gnu.org/software/glpk/glpk.html. |
| Experiment Setup | Yes | We chose n = 2k, because having one manipulator for every two truthful voters represents a sweet-spot where manipulators have enough power to change outcomes, but not too much. ... In our experiments, we chose k m or smaller, because our algorithms are suitable for low k values. ... We performed two sets of experiments, where we drew the non-manipulator votes from either a uniform distribution, or from a P olya-Eggenberger urn model... To model what we see as a relatively realistic setting, we randomly chose weights according to the Zeta distribution... |