Corpus ID: 224804010

Where Is the Normative Proof? Assumptions and Contradictions in ML Fairness Research

@article{Cooper2020WhereIT,
  title={Where Is the Normative Proof? Assumptions and Contradictions in ML Fairness Research},
  author={Andrew Cooper},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.10407}
}
  • Andrew Cooper
  • Published 2020
  • Computer Science
  • ArXiv
  • Across machine learning (ML) sub-disciplines researchers make mathematical assumptions to facilitate proof-writing. While such assumptions are necessary for providing mathematical guarantees for how algorithms behave, they also necessarily limit the applicability of these algorithms to different problem settings. This practice is known - in fact, obvious - and accepted in ML research. However, similar attention is not paid to the normative assumptions that ground this work. I argue such… CONTINUE READING
    1 Citations

    References

    SHOWING 1-10 OF 45 REFERENCES
    Mathematical Notions vs. Human Perception of Fairness: A Descriptive Approach to Fairness for Machine Learning
    • 29
    • Highly Influential
    • PDF
    On the (im)possibility of fairness
    • 208
    • PDF
    Human Comprehension of Fairness in Machine Learning
    • 1
    • PDF
    Fair Enough: Improving Fairness in Budget-Constrained Decision Making Using Confidence Thresholds
    • 3
    • PDF
    Is There a Trade-Off Between Fairness and Accuracy? A Perspective Using Mismatched Hypothesis Testing
    • 6
    • Highly Influential
    • PDF
    Active Fairness in Algorithmic Decision Making
    • 28
    • PDF
    Inherent Trade-Offs in the Fair Determination of Risk Scores
    • 659
    • PDF
    Fairness through awareness
    • 1,127
    • PDF
    The cost of fairness in binary classification
    • 114
    • PDF