mlsanityData to 7 Oct 2026

Fair Representations by Compression

arXiv 2105.14044 ↗First posted 28 May 2021cs.LG cs.CYcited by 7

Not on the map. Why?

Abstract (arXiv)

Organizations that collect and sell data face increasing scrutiny for the discriminatory use of data. We propose a novel unsupervised approach to transform data into a compressed binary representation independent of sensitive attributes. We show that in an information bottleneck framework, a parsimonious representation should filter out information related to sensitive attributes if they are provided directly to the decoder. Empirical results show that the proposed method, FBC, achieves state-of-the-art accuracy-fairness trade-off. Explicit control of the entropy of the representation bit stream allows the user to move smoothly and simultaneously along both rate-distortion and rate-fairness curves. endabstract

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 33 of its 48 references that are in the map. First posted on arXiv 28 May 2021.

Builds on

33 of 48 references.

References · 33

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  2. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  3. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  4. 2016Pixel Recurrent Neural Networkscited by 814
  5. 2014Certifying and removing disparate impactcited by 715
  6. 2018Mitigating Unwanted Biases with Adversarial Learningcited by 663
  7. 2018A Reductions Approach to Fair Classificationcited by 585
  8. 2017Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairnesscited by 381
  9. 2018Learning Adversarially Fair and Transferable Representationscited by 353
  10. 2017Lossy Image Compression with Compressive Autoencoderscited by 289
  11. 2016Revisiting Classifier Two-Sample Testscited by 258
  12. 2015The Variational Fair Autoencodercited by 223
21 more
  1. 2017Emergence of Invariance and Disentanglement in Deep Representationscited by 213
  2. 2019Flexibly Fair Representation Learning by Disentanglementcited by 196
  3. 2015Censoring Representations with an Adversarycited by 191
  4. 2018FairGAN: Fairness-aware Generative Adversarial Networkscited by 164
  5. 2017Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representationscited by 149
  6. 2019On the Fairness of Disentangled Representationscited by 144
  7. 2018The Frontiers of Fairness in Machine Learningcited by 139
  8. 2018Conditional Probability Models for Deep Image Compressioncited by 128
  9. 2018Invariant Representations without Adversarial Trainingcited by 111
  10. 2018Learning Controllable Fair Representationscited by 83
  11. 2018Fairness Through Computationally-Bounded Awarenesscited by 55
  12. 2019A Closer Look at the Optimization Landscapes of Generative Adversarial Networkscited by 25
  13. 2018Creating Fair Models of Atherosclerotic Cardiovascular Disease Riskcited by 23
  14. 2017Provably Fair Representationscited by 22
  15. 2019Invariant Representations through Adversarial Forgettingcited by 20
  16. 2018Unsupervised Adversarial Invariancecited by 14
  17. 2020Learning Smooth and Fair Representationscited by 12
  18. 2014Learning unbiased featurescited by 10
  19. 2019Discovery and Separation of Features for Invariant Representation Learningcited by 4
  20. 2019Multi-Differential Fairness Auditor for Black Box Classifierscited by 3
  21. 2018Bounded Information Rate Variational Autoencoderscited by 1

Built on by

Cited by year

Citing papers per year: 2022: 4; 2023: 1; 2024: 1; 2025: 1.

7 papers.

  1. 2024The Pursuit of Fairness in Artificial Intelligence Models: A Surveycited by 13
  2. 2022Is Fairness Only Metric Deep? Evaluating and Addressing Subgroup Gaps in Deep Metric Learningcited by 11
  3. 2022Ultra-marginal Feature Importance: Learning from Data with Causal Guaranteescited by 6
  4. 2022SoFaiR: Single Shot Fair Representation Learningcited by 1
  5. 2022Ex-Ante Assessment of Discrimination in Datasetcited by 0
  6. 2023Learning and DiSentangling Patient Static Information from Time-series Electronic HEalth Record (STEER)cited by 0
  7. 2025Differential Adjusted Parity for Learning Fair Representationscited by 0