Pith. sign in

REVIEW 9 cited by

Projection to Fairness in Statistical Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2005.11720 v4 pith:TNNCY6NY submitted 2020-05-24 cs.LG math.STstat.MLstat.TH

classification cs.LGmath.STstat.MLstat.TH
keywords fairnessaccuracyestimatorpredictionprojectionfairapproachclosest
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In the context of regression, we consider the fundamental question of making an estimator fair while preserving its prediction accuracy as much as possible. To that end, we define its projection to fairness as its closest fair estimator in a sense that reflects prediction accuracy. Our methodology leverages tools from optimal transport to construct efficiently the projection to fairness of any given estimator as a simple post-processing step. Moreover, our approach precisely quantifies the cost of fairness, measured in terms of prediction accuracy.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Group Fairness Meets the Black Box: Enabling Fair Algorithms on Closed LLMs via Post-Processing

    cs.LG 2025-08 conditional novelty 7.0 of 10

    A prompt-based pipeline lets closed LLMs like GPT-4o be used with classical group-fairness algorithms, without access to weights or embeddings.

  2. OT-FairBoost: Optimal Transport-Guided Gradient Boosting for Fairness Regularization on Tabular Data

    math.ST 2026-07 conditional novelty 6.0 of 10

    Embedding discrete Wasserstein-2 gradients and diagonal Hessians into LightGBM yields stronger accuracy–fairness trade-offs than prior in- and post-processing baselines on classification, regression, and multi-group tasks.

  3. Functional Bilevel Optimization for Predictive Fairness

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Optimizing DPVar—the variance of conditional-mean predictions given a continuous sensitive attribute—via functional bilevel methods (FBO/ITD) yields better fairness–accuracy trade-offs than HSIC, adversarial, and GDP ...

  4. Optimal Transport under Group Fairness Constraints

    stat.ML 2026-01 conditional novelty 6.0 of 10

    Group-fairness targets are added as constraints to entropic optimal transport, with a modified Sinkhorn algorithm and two relaxations (penalty and cost learning) that come with sample-complexity bounds.

  5. Federated Calculation of the Free-Support Transportation Barycenter by Single-Loop Dual Decomposition

    cs.LG 2025-07 conditional novelty 6.0 of 10

    A single-loop dual subgradient method computes a free-support Wasserstein barycenter in a federated setting without solving mass transportation subproblems at each iteration.

  6. Computing Optimal Transport Maps and Wasserstein Barycenters Using Conditional Normalizing Flows

    stat.ML 2025-05 conditional novelty 6.0 of 10

    A conditional normalizing flow method that solves the primal optimal transport problem and computes Wasserstein-2 barycenters as weighted averages of maps from a shared latent distribution.

  7. Fairness-Aware Grouping for Continuous Sensitive Variables: Application for Debiasing Face Analysis with respect to Skin Tone

    cs.CV 2025-07 conditional novelty 5.0 of 10

    A fairness-based grouping algorithm that partitions a continuous sensitive attribute into subgroups with maximally different discrimination levels, validated on synthetic data and face images.

  8. Distributional Limit Theory for Optimal Transport

    math.ST 2025-05 conditional novelty 5.0 of 10

    A survey of central limit theorems for empirical optimal transport, with a new one-dimensional L1 cost fluctuation CLT and a list of open problems.

  9. Machine Unlearning via Information Theoretic Regularization

    cs.LG 2025-02 reject novelty 5.0 of 10

    The paper introduces an auditable, information-theoretic 'marginal unlearning' definition and a rate-distortion style regularization method that can remove data points or features from models.

Pith tools