A General Approach to Fairness with Optimal Transport

Authors: Chiappa Silvia, Jiang Ray, Stepleton Tom, Pacchiano Aldo, Jiang Heinrich, Aslanides John3633-3640

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate its performance on several benchmark fairness datasets. Overall, our methods reach higher fairness with lower loss in accuracy.
Researcher Affiliation Collaboration 1Deep Mind London, 2UC Berkeley, 3Google Research
Pseudocode Yes Algorithm 1 Wass-p Geodesic
Open Source Code No The paper does not provide any statement or link for open-source code for the described methodology.
Open Datasets Yes We evaluated our approach on the National Longitudinal Survey of Youth (NLSY) 1979 regression dataset, on the UCI Communities & Crime (C&C) (Lichman 2013) regression dataset, on the Law School Admission Council (LSAC) regression dataset, and on the UCI Adult binary classification dataset.
Dataset Splits No The paper mentions 'test performance' and training, but does not explicitly provide details on train/validation/test dataset splits (percentages, counts, or specific predefined splits) within the main text.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup No The paper mentions tuning hyperparameters and defining loss functions, but does not provide specific values for hyperparameters or detailed training configurations in the main text.