Corpus ID: 53961090

Algorithmic Bias : A Counterfactual Perspective

@inproceedings{Cowgill2017AlgorithmicB,
  title={Algorithmic Bias : A Counterfactual Perspective},
  author={Bo Cowgill and C. Tucker},
  year={2017}
}
  • Bo Cowgill, C. Tucker
  • Published 2017
  • We discuss an alternative approach to measuring bias and fairness in machine learning: Counterfactual evaluation. In many practical settings, the alternative to a biased algorithm is not an unbiased one, but another decision method such as another algorithm or human discretion. We discuss statistical techniques necessary for counterfactual comparisons, which enable researchers to quantify relative biases without access to the underlying algorithm or its training data. We close by discussing the… CONTINUE READING
    8 Citations
    Procedural Justice and Risk-Assessment Algorithms
    • 2
    • PDF
    Human Intervention in Automated Decision-Making: Toward the Construction of Contestable Systems
    • 3
    Bias in word embeddings
    • 10
    2 Fairness in Machine Learning : key methodological components
    • 2020
    ORIGIN PRIVACY – PROTECTING PRIVACY IN THE BIG-DATA
    • 3
    • PDF

    References

    SHOWING 1-9 OF 9 REFERENCES
    Algorithmic decision making and the cost of fairness
    • 169
    • PDF
    Fairer and more accurate, but for whom?
    • 30
    • PDF
    Estimating causal effects of treatments in randomized and nonrandomized studies.
    • 6,348
    • PDF
    Automating Judgement and Decisionmaking : Theory and Evidence from Résumé Screening
    • 8
    • PDF
    Towards A Rigorous Science of Interpretable Machine Learning
    • 1,021
    • PDF
    How algorithms impact judicial decisions
    • 2017
    Machine bias: There’s software used across the country to predict future criminals. and it’s biased against blacks
    • 2016