Corpus ID: 166228088

Average Individual Fairness: Algorithms, Generalization and Experiments

@inproceedings{Kearns2019AverageIF,
  title={Average Individual Fairness: Algorithms, Generalization and Experiments},
  author={M. Kearns and A. Roth and Saeed Sharifi-Malvajerdi},
  booktitle={NeurIPS},
  year={2019}
}
  • M. Kearns, A. Roth, Saeed Sharifi-Malvajerdi
  • Published in NeurIPS 2019
  • Computer Science, Mathematics
  • We propose a new family of fairness definitions for classification problems that combine some of the best properties of both statistical and individual notions of fairness. We posit not only a distribution over individuals, but also a distribution over (or collection of) classification tasks. We then ask that standard statistics (such as error or false positive/negative rates) be (approximately) equalized across individuals, where the rate is defined as an expectation over the classification… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Paper Mentions

    BLOG POST
    Preference-Informed Fairness
    5
    Ensuring Fairness Beyond the Training Data
    A Notion of Individual Fairness for Clustering
    Beyond Individual and Group Fairness
    On the Fairness of Randomized Trials for Recommendation With Heterogeneous Demographics and Beyond
    A Fair Classifier Using Mutual Information
    Individual Calibration with Randomized Forecasting
    1

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 18 REFERENCES
    Fairness through awareness
    966
    Probably Approximately Metric-Fair Learning
    31
    Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness
    197
    Fairness in Learning: Classic and Contextual Bandits
    196
    Fairness Through Computationally-Bounded Awareness
    45
    An Empirical Study of Rich Subgroup Fairness for Machine Learning
    44
    Online Learning with an Unknown Fairness Metric
    50
    Inherent Trade-Offs in the Fair Determination of Risk Scores
    586
    A Reductions Approach to Fair Classification
    226
    Meritocratic Fairness for Infinite and Contextual Bandits
    10