Balanced Ranking with Diversity Constraints

Authors: Ke Yang, Vasilis Gkatzelis, Julia Stoyanovich

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we describe the results of an experimental evaluation with three empirical goals: (1) to understand the conditions that lead to reduced in-group fairness; (2) to ascertain the feasibility of mitigating the observed imbalance with the help of in-group fairness constraints; (3) to explore the trade-offs between in-group fairness and utility.
Researcher Affiliation Academia 1New York University, Department of Computer Science and Engineering 2Drexel University, Department of Computer Science
Pseudocode No The paper provides mathematical formulations for integer linear programs but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any statement about releasing source code or provide a link to a code repository for the described methodology.
Open Datasets Yes Medical Expenditure Panel Survey (MEPS) MEPS is a comprehensive source of individual and household-level information regarding the amount of health expenditures by individuals from various demographic or socioeconomic groups [Cohen et al., 2009; Coston et al., 2019].
Dataset Splits No The paper describes the datasets used (MEPS and CS) and how items were selected for the experimental dataset, but it does not specify explicit training, validation, or test splits or their proportions.
Hardware Specification No The paper states that 'standard integer programming libraries allow us to solve reasonably large instances in a very small amount of time' and that the solver would complete 'in a matter of a few minutes' for the most time-consuming case, but it does not provide any specific details about the hardware used (e.g., CPU, GPU models, memory).
Software Dependencies No The paper mentions using 'standard integer programming libraries' but does not specify the names of these libraries or their version numbers.
Experiment Setup No The paper describes the formulation of integer linear programs and the use of the leximin solution, but it does not provide specific experimental setup details such as hyperparameter values, training configurations, or system-level settings.