Shizhou Xu

Shizhou Xu

Working on the interdisciplinary area between mathematics and machine learning.

Fair Data Representation for Machine Learning at the Pareto Frontier

Published in Journal of Machine Learning Research (JMLR), 24 (2023), 1–63, 2023

Summary

This paper studies how to transform data into representations that preserve predictive utility while controlling fairness criteria, and how to do so with provable trade-offs rather than ad-hoc heuristics.

Key contributions

  • A Pareto-frontier formulation for fairness–utility trade-offs in representation learning.
  • Algorithms to compute/trace the frontier and select operating points in a transparent way.
  • Theoretical guarantees characterizing when improvements are possible and how costs scale.

Resources

  • Paper (PDF): https://www.jmlr.org/papers/volume24/22-0005/22-0005.pdf
  • Code: (add link when public)
  • Slides: (optional)

Suggested BibTeX

@article{xu2023paretofrontier,
  title   = {Fair Data Representation for Machine Learning at the Pareto Frontier},
  author  = {Xu, Shizhou and Strohmer, Thomas},
  journal = {Journal of Machine Learning Research},
  volume  = {24},
  pages   = {1--63},
  year    = {2023}
}

Recommended citation: Shizhou Xu, Thomas Strohmer. (2023). “Fair Data Representation for Machine Learning at the Pareto Frontier.” Journal of Machine Learning Research, 24, 1–63.
Download Paper | Download Slides