Who's the (Multi-)Fairest of Them All: Rethinking Interpolation-Based Data Augmentation Through the Lens of Multicalibration

Charumathi Badrinath; Karina Halevy; Karly Hou

REVIEW 4 major objections 5 minor 29 references

Who's the (Multi-)Fairest of Them All: Rethinking Interpolation-Based Data Augmentation Through the Lens of Multicalibration

T0 review · 4 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Fair Mixup only improves multicalibration when tuned for one group at a time, while plain Mixup reliably reduces violations and, paired with multicalibration post-processing, yields average balanced-accuracy and MC-violation gains up to…

desk verdict A solid empirical negative result about Fair Mixup under multicalibration, undermined by a misleading composite metric and sloppy pseudocode in the appendix. read the letter →

arxiv 2412.10575 v2 pith:FHE6T7NI submitted 2024-12-13 cs.LG cs.AIstat.ML

Karina Halevy , Karly Hou , Charumathi Badrinath This is my paper · ORCID

classification cs.LGcs.AIstat.ML

keywords multicalibrationdataaugmentationMixupFairalgorithmicfairnesspost-processingintersectionalgroupstabularclassification

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper uses multicalibration to re-examine whether interpolation-based data augmentation actually improves fairness. Across 80 prediction tasks built from US census data, covering up to 81 intersecting demographic groups, it finds that Fair Mixup usually makes fairness worse: its group-balanced batches and pairwise fairness penalty hurt both accuracy and multicalibration. Plain Mixup, which simply trains on interpolated pairs, consistently reduces multicalibration violations. Combining plain Mixup with multicalibration post-processing gives the largest measured gains. The paper argues that the fairness benefits of Fair Mixup do not extend beyond its original setting of one relatively large minority group.

What carries the argument

The central object is the multicalibration violation, defined over groups and prediction bins as the largest absolute difference between the mean predicted probability and the mean true outcome within each bin. The paper operationalizes this with a discretized observable calibration that splits predictions into intervals. The analysis also relies on a component-wise decomposition of Fair Mixup into balanced batch selection (C1), linear interpolation of training points (C2), a pairwise fairness penalty in the loss (C3), and post-processing via the ENFORCE MA/MC algorithms, which learn additive prediction adjustments from a holdout set. This decomposition lets the authors attribute observed fairness changes to specific mechanisms.

What would settle it

Train a predictor with Mixup on a dataset with small groups, run ENFORCE MC on a 25% holdout, and then evaluate the worst-group MC violation on a genuinely unseen test set; if the test-time violation is not lower than the unpost-processed predictor's, the transfer assumption fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that Fair Mixup's fairness benefits evaporate when evaluated through multicalibration on multiple, intersecting, and often tiny demographic groups. In nearly every experiment, all four Fair Mixup variants (demographic parity, equalized odds, multiaccuracy, and multicalibration penalties) worsen baseline multicalibration and balanced accuracy, with the sole exception of the single-group disability setting on the EMPLOYMENT task. Meanwhile, vanilla Mixup consistently lowers worst-group MC violations in most settings, and the best results come from pairing it with the ENFORCE MC post-processing algorithm, improving average balanced accuracy and MC violation by up to 14.22%. By decomposing Fair Mixup into balanced batch selection, linear interpolation, and a pairwise fairness penalty, the paper identifies interpolation of synthetic points as the component that drives fairness gains, while balanced batch selection and the pairwise penalty detract from performance.

Load-bearing premise

The paper's central comparison assumes that the prediction adjustments learned by multicalibration post-processing on a 25% holdout set also hold for unseen test data, even though the formal guarantees only cover the holdout set.

Editorial extensions

If this is right

Fair Mixup should be expected to help only in single-group, binary classification settings; its group-balancing and pairwise-penalty components can actively worsen multicalibration on many intersecting groups.
Plain Mixup is a robust default for improving multicalibration on tabular classification without large accuracy sacrifices.
Pairing Mixup with multicalibration post-processing beats either approach alone in most settings, suggesting augmentation can substitute for some of the holdout data that post-processing would otherwise consume.
The formal guarantees of MC post-processing apply only to the post-processed holdout set, so test-time improvements remain an empirical claim about transfer, not a proven bound.
Balanced accuracy and worst-group MC violation improvements of up to 14.22% are achievable with Mixup plus ENFORCE MC on the INCOME tasks across many states and years.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The finding suggests that fairness-tuned augmentation objectives may be overfit to the specific fairness metric they optimize, and simpler augmentation may generalize better; this could be tested on image and text domains where Mixup variants are common.
The component analysis implies that pairwise fairness penalties should be avoided when groups are numerous and small; a testable extension is to run Mixup with a global, non-pairwise multicalibration penalty to see if it recovers some of the gains without the degradation.
The transfer problem of MC post-processing could be addressed by evaluating on a second holdout or by combining holdout-learned updates with a small calibration set at test time, which would directly quantify how much distribution shift undermines the reported test-time gains.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Here's my take. The core negative result is real and worth knowing: Fair Mixup, evaluated under multicalibration with up to 81 groups, generally makes things worse, and only helps in the single-group binary setting it was designed for. That is a useful corrective to Chuang and Mroueh's earlier claims. The paper also shows vanilla Mixup consistently reduces MC violations, and combining it with MC post-processing gives the best numbers. The experimental scale is impressive: 40 datasets, five group settings, 13 methods, 10 trials, with code and data released. Credit where due: this is the first proper stress-test of interpolation augmentation under multicalibration with many small groups.

The soft spots are mostly in presentation and reporting. The headline '14.22% improvement' is an average of two incommensurable quantities: percent increase in balanced accuracy (often negative) and percent decrease in MC violation (often large). In the biggest case, balanced accuracy drops about 0.7% while MC alpha drops about 29%, and the average of those is called an 'improvement.' That is misleading. The authors should report the two numbers separately, or use a proper multi-objective analysis. There are also no variance estimates across the 10 trials, so we cannot tell whether differences between methods are meaningful. Hyperparameters were chosen on a single state-year (CA 2022) and then fixed elsewhere; this is disclosed, but it still risks overfitting to one slice.

Listing 1, the MC post-processing pseudocode, has clear bugs: undefined variables (data_sum, S_deciles, l) and a reference to a nonexistent list. The actual code on GitHub may be fine, but the listing as printed cannot be run, and that is sloppy for a methods paper.

On the stress-test concern about post-processing transfer: the paper explicitly applies holdout-learned adjustments to test predictions and reports test-time MC violations. So the transfer is measured, not assumed. The limitation is that for very small groups the holdout contains few examples, making the learned adjustments noisy. But the empirical test evaluation is the right way to check, and the paper does it. The concern is real but not load-bearing.

Who is this for? People working on fairness in tabular classification and multicalibration post-processing. It deserves a serious referee, but the authors need to clean up the reporting: fix the pseudocode, give error bars, report accuracy and MC separately, and be more careful about the composite metric. I would send it to review with the expectation of major revision.

Referee Report

4 major / 5 minor

Summary. The paper investigates whether interpolation-based data augmentation improves fairness measured by multicalibration (MC) rather than binary group-fairness metrics. It introduces several variants of Fair Mixup (FMDP, FMEO, FMMA, FMMC) and of Mixup (MIXUP, MIXUP EO, MIXUP MA, MIXUP MC), plus post-processing methods (ENFORCE MA, ENFORCE MC, MIXUP ENFORCE MC), and evaluates them on employment and income prediction from folktables across 10 states by 4 years (40 datasets per task), considering up to 81 marginalized groups in five group-selection settings. The main findings are that Fair Mixup improves MC fairness only in the single-group setting for which it was designed, vanilla Mixup consistently reduces MC violations relative to baseline, and combining Mixup with MC post-processing yields the largest average improvement in a composite balanced-accuracy/MC-violation metric (up to 14.22%). The authors also provide pseudocode for MC and MA post-processing and release code.

Significance. If the results are reliable, the paper provides a useful stress test of Fair Mixup under multicalibration and multiple small groups, and it challenges the prevailing use of Fair Mixup in multi-group settings. The evaluation is extensive (80 tasks, 10 trials), the negative result on Fair Mixup is clearly documented, and the code is released. The paper is also honest about the limitation that MC post-processing guarantees apply only to the holdout set. However, the central positive claim depends on an unquantified transfer of holdout-learned adjustments to test data, and the reported differences lack variance estimates; both need to be addressed before the conclusions can be considered robust.

major comments (4)

[§4.4, Table 3] The central positive claim—that MIXUP ENFORCE MC yields average balanced-accuracy/MC-violation improvements up to 14.22%—rests on applying additive prediction updates learned on a 25% post-processing holdout to unseen test predictions. The manuscript itself states in Section 1 that the guarantees of Hebert-Johnson et al. (2018) apply only to the post-processed holdout set, not to unseen test data. No holdout-versus-test comparison or transfer-gap quantification is provided. Because the holdout is small for the rarest groups (down to tens of individuals), the learned adjustments may overfit the holdout, and the reported test-time improvements could shrink or reverse under a different split. Please report the MC violations on the post-processing holdout alongside the test results, or otherwise quantify the transfer gap, and discuss the sensitivity of the headline improvement to the post-processing split.
[§6, Tables 3, 7, 8] All results are reported as means over ten trials with no variance, confidence intervals, or significance tests. Many of the differences that drive the paper's conclusions are small relative to the scale of the numbers (e.g., Table 3, EMPLOYMENT ALL: ENFORCE MC 8.31 vs MIXUP ENFORCE MC 10.36; Table 8, EMPLOYMENT DLFR: MIXUP ENFORCE MC 0.484 vs ENFORCE MC 0.486), and some central comparisons, such as MIXUP ENFORCE MC versus ENFORCE MC on DIS, are negative. Without measures of dispersion or paired hypothesis tests across the 40 (state, year) pairs, the claim that Mixup 'consistently' makes predictors fairer is not supported. Please add standard errors or per-seed distributions and run paired tests for the main comparisons.
[Appendix C.3, Listings 1 and 2] The pseudo-code defining the ENFORCE algorithms contains multiple undefined variables and syntax errors. In Listing 1, line 28 accumulates `data_sum` without initializing it; line 18 appends `l` to `S_vs[i]` where `l` is not defined in scope; line 57 iterates over `S_deciles[j]`, which is never defined; and line 58 uses `e` outside its loop context. Listing 2, line 41, has a stray colon after the assignment to `done`. Because these listings specify the exact post-processing procedures whose outputs are the basis for the paper's main empirical claims, please correct them and make them consistent with the released implementation.
[Appendix A.1] The hyperparameters (d, k, λ) were selected on California 2022 and then applied to the other 39 state-year subsets. The paper acknowledges this, but the aggregated results in Table 3 therefore include 39 datasets for which the hyperparameters were not tuned, and overfitting to CA 2022 could bias the averages. Please report the results with CA 2022 excluded, or discuss the sensitivity of the conclusions to the hyperparameter choice.

minor comments (5)

[§6, Table 3] Each entry in Table 3 appears to be the mean of two different quantities (percent increase in balanced accuracy and percent decrease in worst-group MC α) across 40 state-year pairs. Please state this explicitly in the table caption and in the text, since collapsing the two metrics into a single number makes it difficult to separate accuracy changes from fairness changes.
[§5.1, Table 2] The notation 'Max # Minority Groups' and 'Max # Groups ≤ 0.25% of Population' is not fully defined; please clarify that these maxima are taken over the 40 state-year subsets for each task.
[§4.4] The description of ENFORCE MA says it uses Algorithm 3.1 of Hebert-Johnson et al. (2018), but Listing 2 implements a similar routine; please state whether the released code follows Listing 2 exactly and note any differences from the original algorithm.
[§2.3, Eq. (2)] For the discretization in Eq. (2), the last interval for v = d is [1, 1 + 1/d) but probabilities are in [0,1]; please specify how the boundary fi = 1 is handled.
[Header] The paper header includes both an arXiv number and a 'Proceedings version' link to an AAAI article; please clarify in the submission the relationship between this manuscript and the AAAI proceedings version, including what content is new.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims are empirical comparisons on held-out test metrics, and the post-processing transfer limitation is explicitly disclosed rather than hidden.

full rationale

This paper is an empirical benchmark study rather than a derivation, and its central comparisons do not reduce to their inputs by construction. The headline results in Table 3 are test-time balanced accuracy and worst-group multicalibration (MC) violation computed on held-out test data. For ENFORCE MC and MIXUP ENFORCE MC, the prediction-adjustment rules are learned on a 25% holdout set and then applied to test predictions, which makes the test MC numbers an empirical generalization measurement rather than a fitted quantity. The MC and multiaccuracy penalties used during training in FMMC and FMMA are the same notions later evaluated, but this is objective alignment, not circularity, because the reported violations are measured on test data not used to fit those penalties. Hyperparameters were selected once on California 2022 and applied to the other 39 state-year datasets, so the aggregate improvements are not forced by construction. The paper explicitly notes that the guarantees of Hebert-Johnson et al. (2018) apply only to the post-processed holdout set and not to unseen test data, which is a disclosed limitation about transfer, not a circular dependency. No load-bearing self-citations, uniqueness arguments, or ansatz-smuggling citations appear; the Fair Mixup methods are taken from prior work and then stress-tested on new data and metrics. Therefore no circular step can be exhibited, and the appropriate finding is no significant circularity.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central empirical claims rest on the chosen hyperparameters (d, k, lambda, alpha) and on the transfer of post-processing updates to test data. The MC definition itself is modified for computational tractability.

free parameters (6)

d (MC discretization) = 10 (also tried 55, 100)
Selected on CA 2022; controls resolution of prediction intervals in FMMC and ENFORCE MC.
k (number of group gradients averaged) = 3, 40, or 100 by method (Table 6)
Selected on CA 2022; controls how many groups contribute to the multi-group fairness penalty.
lambda (fairness penalty weight) = 0.25 or 0.5 by method (Table 6)
Selected on CA 2022; controls strength of fairness penalty relative to BCE loss.
alpha (post-processing violation target) = 0.01
Chosen to be smaller than the smallest mean MC violation among post-processing-free methods, so the post-processor always makes updates.
batch sampling threshold = 183617.4
Threshold on group size for choosing sampling with replacement; determined from CA 2022 x ALL experiments.
epsilon (Beta distribution parameter) = 1
Fixed following Chuang and Mroueh (2021); controls mixup interpolation strength.

assumptions (4)

domain assumption Binary labels are used to estimate true probabilities: yi replaces p*_i in calibration calculations.
Needed to compute multicalibration from classification outcomes; standard in observable calibration (Hebert-Johnson et al. 2018, Eq. 1).
domain assumption The folktables ACS dataset provides a reasonable and sufficiently unbiased testbed for fairness methods.
Conclusions about augmentation methods are drawn from income and employment tasks across 40 US state-year subsets; representativeness of these datasets is assumed.
ad hoc to paper MC post-processing adjustments learned on a holdout set can be applied to test predictions.
The paper augments the algorithm with update rules for unseen data (Section 4.4), though the original theory only guarantees enforcement on the holdout set (Section 1).
ad hoc to paper Discretized multicalibration requiring S=S' is a valid working definition.
The paper modifies the MC definition to require calibration on all of S and discretizes predictions (Section 2.3), deviating from Hebert-Johnson et al.'s 1-alpha subset condition.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Who's the (Multi-)Fairest of Them All: Rethinking Interpolation-Based Data Augmentation Through the Lens of Multicalibration." pith.science (2026). https://pith.science/paper/FHE6T7NI

@misc{pith2026241210575,
  author       = {Pith},
  title        = {Pith review of: Who's the (Multi-)Fairest of Them All: Rethinking Interpolation-Based Data Augmentation Through the Lens of Multicalibration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FHE6T7NI}},
  note         = {Machine review of arXiv:2412.10575}
}

read the original abstract

Data augmentation methods, especially SoTA interpolation-based methods such as Fair Mixup, have been widely shown to increase model fairness. However, this fairness is evaluated on metrics that do not capture model uncertainty and on datasets with only one, relatively large, minority group. As a remedy, multicalibration has been introduced to measure fairness while accommodating uncertainty and accounting for multiple minority groups. However, existing methods of improving multicalibration involve reducing initial training data to create a holdout set for post-processing, which is not ideal when minority training data is already sparse. This paper uses multicalibration to more rigorously examine data augmentation for classification fairness. We stress-test four versions of Fair Mixup on two structured data classification problems with up to 81 marginalized groups, evaluating multicalibration violations and balanced accuracy. We find that on nearly every experiment, Fair Mixup \textit{worsens} baseline performance and fairness, but the simple vanilla Mixup \textit{outperforms} both Fair Mixup and the baseline, especially when calibrating on small groups. \textit{Combining} vanilla Mixup with multicalibration post-processing, which enforces multicalibration through post-processing on a holdout set, further increases fairness.

Figures

Figures reproduced from arXiv: 2412.10575 by the authors.

**Figure 1.** The ML training and evaluation pipelines considered in our work. Each method in our experiments can be character [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Venn Diagram of each method’s core components. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 24 canonical work pages

[1]

Allison, K.; Patel, D.; and Kaur, R. 2022. Assessing multiple factors affecting minority participation in clinical trials: Development of the clinical trials participation barriers survey. Cureus, 14(4): e24424

work page 2022
[2]

N.; Shalit, U.; Netzer, D.; Balicer, R.; and Dagan, N

Barda, N.; Riesel, D.; Akriv, A.; Levy, J.; Finkel, U.; Yona, G.; Greenfeld, D.; Sheiba, S.; Somer, J.; Bachmat, E.; Rothblum, G. N.; Shalit, U.; Netzer, D.; Balicer, R.; and Dagan, N. 2020. Developing a COVID -19 mortality risk prediction model when individual-level data are not available. Nature Communications, 11(1): 4439

work page 2020
[3]

Becker, B.; and Kohavi, R. 1996. Census Income Data Set

work page 1996
[4]

T.; and Nakkiran, P

B asiok, J.; Gopalan, P.; Hu, L.; Kalai, A. T.; and Nakkiran, P. 2024. Loss Minimization Yields Multicalibration for Large Neural Networks. In Guruswami, V., ed., 15th Innovations in Theoretical Computer Science Conference, ITCS 2024, January 30 to February 2, 2024, Berkeley, CA, USA , volume 287 of LIPIcs, 17:1--17:21. Schloss Dagstuhl - Leibniz-Zentrum ...

work page 2024
[5]

B asiok, J.; Gopalan, P.; Hu, L.; and Nakkiran, P. 2023. When Does Optimizing a Proper Loss Yield Calibration? In Thirty-seventh Conference on Neural Information Processing Systems

work page 2023
[6]

V.; Bowyer, K

Chawla, N. V.; Bowyer, K. W.; Hall, L. O.; and Kegelmeyer, W. P. 2002. SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16: 321--357

2002
[7]

Chouldechova, A. 2017. Fair Prediction with Disparate Impact: A Study of Bias in Recidivism Prediction Instruments. Big Data, 5(2): 153--163. PMID: 28632438

work page 2017
[8]

Chuang, C.-Y.; and Mroueh, Y. 2021. Fair Mixup: Fairness via Interpolation. In International Conference on Learning Representations

work page 2021

Show all 29 references

[9]

Ding, F.; Hardt, M.; Miller, J.; and Schmidt, L. 2021. Retiring Adult: New Datasets for Fair Machine Learning. In Beygelzimer, A.; Dauphin, Y.; Liang, P.; and Vaughan, J. W., eds., Advances in Neural Information Processing Systems

2021
[10]

J.; Ramdas, A.; and Tibshirani, R

Foygel Barber, R.; Candès, E. J.; Ramdas, A.; and Tibshirani, R. J. 2020. The limits of distribution-free conditional predictive inference . Information and Inference: A Journal of the IMA, 10(2): 455--482

2020
[11]

Google. 2024. Data Commons

2024
[12]

Gopalan, P.; Okoroafor, P.; Raghavendra, P.; Shetty, A.; and Singhal, M. A. 2024. Omnipredictors for Regression and the Approximate Rank of Convex Functions. In COLT

2024
[13]

N.; Lee, M

Guo, L. N.; Lee, M. S.; Kassamali, B.; Mita, C.; and Nambudiri, V. E. 2022. Bias in, bias out: Underreporting and underrepresentation of diverse skin types in machine learning research for skin cancer detection-A scoping review. J. Am. Acad. Dermatol., 87(1): 157--159

2022
[14]

Hansen, D.; Devic, S.; Nakkiran, P.; and Sharan, V. 2024. When is Multicalibration Post-Processing Necessary? arXiv:2406.06487

2024 arXiv
[15]

Hardt, M.; Price, E.; and Srebro, N. 2016. Equality of opportunity in supervised learning. In Proceedings of the 30th International Conference on Neural Information Processing Systems, NIPS'16, 3323–3331. Red Hook, NY, USA: Curran Associates Inc. ISBN 9781510838819

2016
[16]

Hebert-Johnson, U.; Kim, M.; Reingold, O.; and Rothblum, G. 2018. Multicalibration: Calibration for the ( C omputationally-Identifiable) Masses. In Dy, J.; and Krause, A., eds., Proceedings of the 35th International Conference on Machine Learning, volume 80 of Proceedings of M...

2018
[17]

Huang, J.; Galal, G.; Etemadi, M.; and Vaidyanathan, M. 2022. Evaluation and mitigation of racial bias in clinical machine learning models: Scoping review. JMIR Med. Inform., 10(5): e36388

2022
[18]

Iosifidis, V.; and Ntoutsi, E. 2018. Dealing with Bias via Data Augmentation in Supervised Learning Scenarios. Jo Bates Paul D. Clough Robert J \"a schke , 24

2018
[19]

Jung, C.; Lee, C.; Pai, M.; Roth, A.; and Vohra, R. 2021. Moment Multicalibration for Uncertainty Estimation. In Belkin, M.; and Kpotufe, S., eds., Proceedings of Thirty Fourth Conference on Learning Theory, volume 134 of Proceedings of Machine Learning Research, 2634--2678. PMLR

2021
[20]

Jung, C.; Noarov, G.; Ramalingam, R.; and Roth, A. 2023. Batch Multivalid Conformal Prediction. In International Conference on Learning Representations

2023
[21]

P.; Ghorbani, A.; and Zou, J

Kim, M. P.; Ghorbani, A.; and Zou, J. 2019. Multiaccuracy: Black-Box Post-Processing for Fairness in Classification. In Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, AIES '19, 247–254. New York, NY, USA: Association for Computing Machinery. ISBN 9781450363242

2019
[22]

Kumar, A.; Sarawagi, S.; and Jain, U. 2018. Trainable Calibration Measures for Neural Networks from Kernel Mean Embeddings. In Dy, J.; and Krause, A., eds., Proceedings of the 35th International Conference on Machine Learning, volume 80 of Proceedings of Machine Learning Resea...

2018
[23]

M.; Bouneffouf, D.; Muthusamy, V.; and Varshney, K

Sharma, S.; Zhang, Y.; R\' os Aliaga, J. M.; Bouneffouf, D.; Muthusamy, V.; and Varshney, K. R. 2020. Data Augmentation for Discrimination Prevention and Bias Disambiguation, 358–364. New York, NY, USA: Association for Computing Machinery. ISBN 9781450371100

2020
[24]

Wadhwa, M.; Bhambhani, M.; Jindal, A.; Sawant, U.; and Madhavan, R. 2022. Fairness for Text Classification Tasks with Identity Information Data Augmentation Methods

2022
[25]

Wald, Y.; Feder, A.; Greenfeld, D.; and Shalit, U. 2021. On Calibration and Out-of-Domain Generalization. In Beygelzimer, A.; Dauphin, Y.; Liang, P.; and Vaughan, J. W., eds., Advances in Neural Information Processing Systems

2021
[26]

A.; and Breckon, T

Yucer, S.; Akcay, S.; Moubayed, N. A.; and Breckon, T. 2020. Exploring racial bias within face recognition via per-subject adversarially-enabled data augmentation. In Computer Vision and Pattern Recognition Workshops. IEEE. To be presented at the Workshop on Fair, Data Efficie...

2020
[27]

N.; and Lopez - Paz, D

Zhang, H.; Ciss \' e , M.; Dauphin, Y. N.; and Lopez - Paz, D. 2017. mixup: Beyond Empirical Risk Minimization. CoRR, abs/1710.09412

2017 arXiv
[28]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...
[29]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

Allison, K.; Patel, D.; and Kaur, R. 2022. Assessing multiple factors affecting minority participation in clinical trials: Development of the clinical trials participation barriers survey. Cureus, 14(4): e24424

work page 2022

[2] [2]

N.; Shalit, U.; Netzer, D.; Balicer, R.; and Dagan, N

Barda, N.; Riesel, D.; Akriv, A.; Levy, J.; Finkel, U.; Yona, G.; Greenfeld, D.; Sheiba, S.; Somer, J.; Bachmat, E.; Rothblum, G. N.; Shalit, U.; Netzer, D.; Balicer, R.; and Dagan, N. 2020. Developing a COVID -19 mortality risk prediction model when individual-level data are not available. Nature Communications, 11(1): 4439

work page 2020

[3] [3]

Becker, B.; and Kohavi, R. 1996. Census Income Data Set

work page 1996

[4] [4]

T.; and Nakkiran, P

B asiok, J.; Gopalan, P.; Hu, L.; Kalai, A. T.; and Nakkiran, P. 2024. Loss Minimization Yields Multicalibration for Large Neural Networks. In Guruswami, V., ed., 15th Innovations in Theoretical Computer Science Conference, ITCS 2024, January 30 to February 2, 2024, Berkeley, CA, USA , volume 287 of LIPIcs, 17:1--17:21. Schloss Dagstuhl - Leibniz-Zentrum ...

work page 2024

[5] [5]

B asiok, J.; Gopalan, P.; Hu, L.; and Nakkiran, P. 2023. When Does Optimizing a Proper Loss Yield Calibration? In Thirty-seventh Conference on Neural Information Processing Systems

work page 2023

[6] [6]

V.; Bowyer, K

Chawla, N. V.; Bowyer, K. W.; Hall, L. O.; and Kegelmeyer, W. P. 2002. SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16: 321--357

2002

[7] [7]

Chouldechova, A. 2017. Fair Prediction with Disparate Impact: A Study of Bias in Recidivism Prediction Instruments. Big Data, 5(2): 153--163. PMID: 28632438

work page 2017

[8] [8]

Chuang, C.-Y.; and Mroueh, Y. 2021. Fair Mixup: Fairness via Interpolation. In International Conference on Learning Representations

work page 2021