Tuning Fairness by Marginalizing Latent Target Labels
Explore this paper's citation graph
Summary
It is shown that bias in the output can naturally be handled in probabilistic models by introducing a latent target output that will modulate the likelihood function, and is expressed as marginalization instead of constrained problems.
- Type
- article
- Published
- 2018-10-12
- Cited by
- 4
- References
- 22
- Access
- Open access
- OpenAlex
- https://openalex.org/W2936224566
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:145966270
Keywords
Business, Law and economics, Computer science, Political science, Sociology
References
- An Introduction to Variational Methods for Graphical Models
- Big Data's Disparate Impact
- Certifying and Removing Disparate Impact
- On the Convergence of the Concave-Convex Procedure
- Learning Fair Representations
- Scalable Inference for Gaussian Process Models with Black-Box Likelihoods
- Satisfying Real-world Goals with Dataset Constraints
- Generic Inference in Latent Gaussian Process Models
- Fair Prediction with Disparate Impact: A Study of Bias in Recidivism Prediction Instruments
- A statistical framework for fair predictive algorithms
- A comparative study of fairness-enhancing interventions in machine learning
- GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration
- Fair and Unbiased Algorithmic Decision Making: Current State and Future Challenges
- Discovering Fair Representations in the Data Domain
- AutoGP: Exploring the Capabilities and Limitations of Gaussian Process Models
- Inherent Trade-Offs in the Fair Determination of Risk Scores
- The Variational Fair Autoencoder
- Scalable Variational Gaussian Process Classification
- Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goals
- Adam: A Method for Stochastic Optimization
Cited by
- SensitiveNets: Learning Agnostic Representations with Application to Face Recognition
- Attraction-Repulsion clustering with applications to fairness
- Tuning Fairness by Balancing Target Labels
- Bias Mitigation for Machine Learning Classifiers: A Comprehensive Survey
- Künstliche Intelligenz in der Hochschulbildung
- Künstliche Intelligenz in der Hochschulbildung