Research map / Learning theory and algorithms / Fairness
Fairness auditing and regularization
Audits and mitigates unfairness with sensitivity analysis, data influence and differentiable regularizers.
65 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2018–2026) · an approach in Fairness, keyword cluster “decision · making · systems”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
23 papers in 2023–24 and 20 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 13.8% to 14.3%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2018 · Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer (NeurIPS 2018 · 56 citations)
- 2019 · Balancing Efficiency and Fairness in On-Demand Ridesourcing (NeurIPS 2019 · 45 citations)
- 2020 · Collaborative Machine Learning with Incentive-Aware Model Rewards (ICML 2020 · 31 citations)
- 2021 · Tilted Empirical Risk Minimization (ICLR 2021 · 38 citations)
- 2022 · Fairness without Imputation: A Decision Tree Approach for Fair Prediction with Missing Values (AAAI 2022 · 24 citations)
- 2023 · Social Bias Meets Data Bias: The Impacts of Labeling and Measurement Errors on Fairness Criteria (AAAI 2023 · 8 citations)
- 2024 · CURE4Rec: A Benchmark for Recommendation Unlearning with Deeper Influence (NeurIPS 2024 · 7 citations)
- 2025 · FairTP: A Prolonged Fairness Framework for Traffic Prediction (AAAI 2025 · 3 citations)
- 2026 · Addressing Polarization and Unfairness in Performative Prediction (AAAI 2026 · 1 citation)
Earliest papers here
- Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer (NeurIPS 2018 · 56 citations)
- Balancing Efficiency and Fairness in On-Demand Ridesourcing (NeurIPS 2019 · 45 citations)
- Bayesian Fairness (AAAI 2019 · 20 citations)
What has been tried
47 papers here report 395 measured changes (a component added, removed or swapped, compared with a baseline): helped 50%, no clear effect 12%, hurt 28%, mixed 10%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add RSF loss | HK · RSF | 1.01 -> 0.91 | Helped | FairTP: A Prolonged Fairness Framework for Traffic Prediction (AAAI 2025 · 3 citations) |
| Limit projection optimization to at most 10 iterations instead of using more than 10 | ACSIncome DP · fairness violation | Helped | fairret: a Framework for Differentiable Fairness Regularization Terms (ICLR 2024 · 1 citation) | |
| Perform selected computations in an offline phase | Fairness certificate proof generation · online proof generation time | largest being 1.74\times | Helped | FairProof : Confidential and Certifiable Fairness for Neural Networks (ICML 2024 · 1 citation) |
| Add L1 regularization with weight 2·10^-6 | MNIST 0.1 · certified accuracy | 98.21 -> 98.26 | Helped | CTBench: A Library and Benchmark for Certified Training (ICML 2025 · 0 citations) |
| Set β=0.7 and γ=1 instead of β=1 and γ=0.5 | Friedman dataset, party 1 versus party 2 · reward-value difference | Mixed | Incentivizing Time-Aware Fairness in Data Sharing (NeurIPS 2025 · 0 citations) | |
| Use a partial weighted lottery with k'=0.5·k and n'=k instead of top-k selection | Normally distributed claims · expected utility | < 0.05 | Hurt | Position: Scarce Resource Allocations That Rely On Machine Learning Should Be Randomized (ICML 2024) |
What reviewers will expect
Datasets and baselines named in the 9 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- MNIST · 2 of 9 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 43 papers; from OpenAlex and the papers' affiliations).
- Inria · 3 papers
- University of Alberta · 2 papers
- University of California, San Diego · 2 papers
- Ludwig-Maximilians-Universität München · 2 papers
- LMU Munich · 2 papers
- Harvard University · 2 papers
- Massachusetts Institute of Technology · 2 papers
- Stanford University · 2 papers
- ETHZ - ETH Zurich · 2 papers
- Swiss Federal Institute of Technology · 2 papers
Use it
Download BibTeX of 37 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Causal and policy fairness in decisions (75 papers)
- Fairness in recommendation and ranking (20 papers)
Elsewhere in the topic
- Fairness-constrained classification (148 papers)
- Post-processing and transport for fairness (101 papers)
- Fair allocation and selection (69 papers)
- Minimax fairness via sampling (38 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).