Skip to main content

> ML_LITERATURE // HARDT-2016-EQUALITY-OF-OPPORTUNITY-IN-SUPERVISED-LEARNING_v1.0

Equality of Opportunity in Supervised Learning

Moritz Hardt, Eric Price, Nathan Srebro · Advances in Neural Information Processing Systems (NeurIPS) (2016)

safety-fairness2016industry-standardartifactsAvailable
arXiv:1610.02413Back to Literature

Principal Contribution

Defined the mathematical fairness criterion "Equality of Opportunity", requiring equal true positive rates (equalized odds) across protected demographic groups.

Operational Relevance

Serves as qualified reference for implementing task-fairness-audit, task-binary-classification in production systems.

Assumptions

  • Underlying computational topology and mathematical bounds adhere to established convexity/smoothness guarantees

Limitations

  • Hardware runtime speedups, privacy budgets, and convergence depend on hyperparameters and network communication limits

Connected Algorithms, Architectures & Tools

Related Algorithms:
Related Architectures:
Implementing Libraries: