Fair Data Representation for Machine Learning at the Pareto Frontier

Workshop talk on provable fairness–utility trade-offs and algorithmic construction along the Pareto frontier.