Retiring Adult: New Datasets for Fair Machine Learning

Authors: Frances Ding, Moritz Hardt, John Miller, Ludwig Schmidt

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We highlight a broad initial sweep of new empirical insights relating to trade-offs between fairness criteria, performance of algorithmic interventions, and the role of distribution shift based on our new datasets.
Researcher Affiliation Collaboration Frances Ding UC Berkeley Moritz Hardt UC Berkeley John Miller UC Berkeley Ludwig Schmidt Toyota Research Institute
Pseudocode No The paper does not contain any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes The datasets and Python package are available for download at https://github.com/zykls/ folktables.
Open Datasets Yes The datasets and Python package are available for download at https://github.com/zykls/ folktables.
Dataset Splits Yes The UCI Machine Learning Repository contributed to this development by providing researchers with numerous datasets each with a fixed training and testing split. ... Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes] Please see Appendix C.
Hardware Specification Yes Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes] Please see Appendix C.
Software Dependencies No The paper mentions a 'Python package called folktables' and various machine learning methods, but does not specify key software components with their version numbers required for reproduction.
Experiment Setup Yes Did you specify all the training details (e.g., data splits, hyperparameters, how they were chosen)? [Yes] Please see Appendix C.