unconstrainedml

Reproduce experiments in "The Implicit Fairness Criterion of Unconstrained Learning" (ICML 2019)

Instructions

1. Clone this repository.
2. Download adult.data.txt and adult.test.txt from the UCI Adult Dataset to the root folder.
3. Download BROWARD_CLEAN.csv from the Broward County Dataset to the root folder.
4. Run all cells in adult.ipynb and broward.ipynb with Python 2.7.15.

Citation

Lydia T. Liu, Max Simchowitz, Moritz Hardt. The implicit fairness criterion of unconstrained learning. Proceedings of the 36th International Conference on Machine Learning (ICML), Long Beach, CA, USA, 2019.

References for datasets

• Dheeru Dua and Efi Karra Taniskidou. UCI machine learning repository, 2017. URL http://archive.ics.uci.edu/ml
• J. Dressel, H. Farid, The accuracy, fairness, and limits of predicting recidivism. Sci. Adv. 4, eaao5580 (2018)

License

This repository is licensed under the BSD 3-Clause "New" or "Revised" License.