Pith. sign in

REVIEW 5 major objections 5 minor 40 references

Achieving Distributive Justice in Federated Learning via Uncertainty Quantification

T0 review · 5 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read The paper claims that one federated learning objective, UDJ-FL, can deliver four distributive-justice fairness ethics—egalitarian, utilitarian, Rawls' difference principle, and desert—by choosing three hyperparameters.

desk verdict Good idea, broken derivation: the claim that one objective realizes all four justice notions doesn't survive arithmetic. read the letter →

arxiv 2504.15924 v1 pith:P6SXAF2I submitted 2025-04-22 cs.LG cs.AIstat.ML

Alycia Carey , Xintao Wu This is my paper · ORCID
classification cs.LGcs.AIstat.ML
keywords federatedlearningclient-levelfairnessdistributivejusticealeatoricuncertaintyquantificationfairresourceallocationhyperparameterselectionsoftmaxentropy
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

This paper tries to establish that a single federated learning objective, UDJ-FL, can produce four different distributive-justice fairness behaviors—egalitarian, utilitarian, Rawls' difference principle, and desert—merely by selecting hyperparameters. The key move is to define each client's advantage and contribution by its average softmax entropy over its own local data, an aleatoric-uncertainty estimate, and then reweight or power-transform client losses with those entropy scores. On two image datasets with five clients of controlled data ambiguity, the paper reports that UDJ-FL attains the intended fairness behavior for all four settings and matches or beats existing fair federated learning methods on the corresponding fairness metrics. If this holds, a practitioner could change the ethical principle governing a federation without changing the training architecture.

What carries the argument

The load-bearing object is the UDJ-FL objective of Eq. 8, $h(\theta)=\sum_i \upsilon_i^{\gamma}H_i(\theta)^{r\beta}$, coupled to the client weight of Eq. 9, $\upsilon_i = \frac{1}{|D_i^{tr}|}\sum_{x\in D_i^{tr}} H(Y|x,\theta)$, the average softmax entropy over the client's local training data. The hyperparameter settings in Table 2 select the fairness theory: $\gamma$ controls whether high-entropy or low-entropy clients get emphasized, $r$ sets the growth rate, and $\beta$ controls the concave or power shape of the loss transformation, with $\beta\to\infty$ for Rawls' principle, $\beta\to 0$ for utilitarian, $\beta=1$ for egalitarian, and client-specific negative exponents for desert. The same machinery also supplies the optimization rule: following $q$-FedAvg, the server aggregates client updates using a local Lipschitz upper bound (Lemma 1) to handle the exponentiated losses, and Lemma 2 uses Hölder's inequality to bound generalization for the uncertainty-weighted objective.

What would settle it

On a synthetic federation, flip labels for the client with the highest true aleatoric uncertainty so that the trained model becomes confidently wrong on that client's data; if UDJ-FL's computed $\upsilon_i$ stays low and the desert correlation $r_{\boldsymbol{\upsilon},A}$ still approaches $-1$, the entropy measure misidentifies contribution and the claimed fairness behavior is an artefact of the metric rather than the method.

Watch

Extended reading notes

Core claim

The paper's central claim is that one federated objective, $$h(\$\theta$) = \sum_{i=1}^{N} \$upsilon_i^{{\gamma}}$ H_i(\$\theta$)^{r\$\beta$},$$ where $H_i(\theta)$ is client $i$'s empirical loss and $\upsilon_i$ is the average softmax entropy of client $i$'s local training data, recovers four distributive-justice fairness ethics purely through the hyperparameters $r$, $\beta$, and $\gamma$. Setting $r=1+1/\beta$, $\gamma=1$, and letting $\beta\to\infty$ yields Rawls' difference principle; $r=1+1/\beta$, $\beta\to 0$, $\gamma=-1$ yields utilitarianism; $r=1$, $\beta=1$, $\gamma=1$ yields egalitarianism; and $r=1$, $\gamma=0$ with client-specific exponents $\beta_i = (1/\upsilon_i)/\sum_j (1/\upsilon_j)$ yields desert-based fairness. The authors argue that a client's aleatoric uncertainty is irreducible and stable, so it is the right quantity for deciding who is least advantaged and who contributes most, unlike dataset size or participation counts in cross-silo settings. On Dirty-MNIST and CURE-TSR with five clients of controlled data ambiguity, UDJ-FL reports the desired fairness behavior for each setting and matches or improves on $q$-FFL, AFL, PropFair, TERM, FedMGDA+, and CFFL on the corresponding fairness measures. The paper also adapts the $q$-FedAvg Lipschitz step-size argument to handle the exponentiated losses and proves a generalization bound for the uncertainty-weighted objective by extending the AFL bound.

Load-bearing premise

The load-bearing premise is that a client's average softmax entropy over its own local training data is a valid, stable measure of that client's data quality, and therefore of its advantage and contribution; if noisy or adversarial local data produces misleadingly confident predictions, the fairness guarantees built on those entropy weights collapse.

Editorial extensions

If this is right

  • A practitioner can select a fairness ethics for a cross-silo federation by setting $r$, $\beta$, and $\gamma$; no change to the learning architecture is needed.
  • UDJ-FL contains $q$-FFL as a special case when $r=1+1/\beta$ and $\beta>0$, so results transfer to that popular baseline's settings.
  • Client contribution can be measured by data quality instead of dataset size or participation count, which matters when the largest client has the noisiest data.
  • The fairness metrics used for evaluation become decoupled from the training objective, so the same trained model can be audited under egalitarian, utilitarian, Rawlsian, or desert criteria.
  • The generalization bound extends the AFL-style analysis to uncertainty-weighted federated objectives, giving a theoretical guarantee for the reweighted loss family.

Reading between the lines

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

  • The same entropy-weighting recipe could be applied to other aggregation rules such as FedProx or SCAFFOLD, since nothing in the argument binds it to averaging-based updates; the paper does not test this.
  • If entropy scores are cheap to compute in one warm-up round, UDJ-FL could double as a diagnostic: comparing outcomes under the four ethics would reveal which clients are data-poor and which are data-rich, a use the authors mention only implicitly.
  • The empirical finding that $\beta=0.1$ beats $\beta=0$ for utilitarian fairness on Dirty-MNIST suggests the pure utilitarian objective of Eq. 12 may not be the practical optimum; a systematic sweep of $\gamma$ and small $\beta$ values is the implied but untested extension.
  • A direct comparison of aleatoric versus epistemic or predictive uncertainty weights would test the paper's irreducibility argument, since the authors do not run that experiment.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 5 minor

Summary. This paper introduces UDJ-FL, a federated learning objective intended to realize four client-level fairness notions drawn from distributive justice—egalitarian, utilitarian, Rawls' difference principle, and desert—through different settings of three hyperparameters r, β, and γ. The client weights are aleatoric uncertainties estimated as the average softmax entropy of each client's training data (Eq. 9). The objective is written in Eq. 8 as sum_i υ_i^γ H_i(θ)^{rβ}, with derivations presented for each fairness notion in Section 4 and summarized in Table 2. The paper contributes Algorithm 1, a q-FedAvg-style solver, a Lipschitz-constant lemma, generalization bounds in Appendix D, and experiments on Dirty-MNIST and CURE-TSR with five clients whose data quality is controlled via clean/ambiguous shards.

Significance. If the algebraic route from Eq. 8 to the four objectives were valid, UDJ-FL would be a practically attractive contribution: one training routine, no architectural changes, hyperparameter selection of the fairness ethics, public code, and experiments against a broad set of baselines. The taxonomy of prior work along the four justice dimensions is also useful. However, the central derivation currently contains internal inconsistencies in the Rawls and desert cases, and the uncertainty weighting in Eq. 9 has a sign issue; these undermine the advertised 'one objective, four fairness notions' claim as written. The empirical results may still be meaningful, but they cannot be fully interpreted until the objectives being executed are identified unambiguously. The public code and the controlled uncertainty-partition experimental setup are concrete strengths.

major comments (5)
  1. [§4.1, Eq. (8)] The step from Lan et al.'s Eq. (5) to Eq. (8) is not derived. Eq. (5) has the exponent −rβ, while Eq. (8) uses +rβ, and the per-client weights υ_i^γ appear without any formal rule connecting q_i in Eq. (5) to υ_i^γ. The sentence that the sign change is due to federated learning minimizing costs rather than distributing utilities explains the sign convention but does not show that the resulting expression belongs to the same fairness family or is an equivalent transformation. Since every row of Table 2 is claimed to be a special case of Eq. (8), this missing equivalence is load-bearing.
  2. [§4.2, Eq. (10) and Table 2] Setting r=1+1/β and γ=1 in Eq. (8) yields sum_i υ_i H_i(θ)^{1+β}. Eq. (10), as typeset, contains a prefactor (1+β) and a weight on υ_i that appears to be υ_i^{1+β}; if the intended weight is instead υ_i, the equation still is not a direct special case of Eq. (8). Thus the Rawls row of Table 2 is not obtained from Eq. (8) by the stated substitutions, and the claim that β→∞ recovers min-max fairness needs a corrected intermediate objective.
  3. [§4.4, Eqs. (13)–(14)] The desert objective is not a special case of Eq. (8). Eq. (13) defines β_i>0, yet Eq. (14) minimizes (1/N) sum_i H_i^{−β_i}, whose monotonicity in H_i is opposite to that of sum_i H_i^{β_i}, which is what Eq. (8) gives for r=1, scalar β=β_i, and γ=0. The text says 'we let β≤0', but Eq. (13) assigns positive values. Moreover, Eq. (8) has a single scalar β, so the per-client vector β cannot be inserted without changing the objective family. This invalidates the desert entry of Table 2 as written.
  4. [§3.3/§4.1, Eq. (9)] Eq. (9) defines υ as the average of p log p over the softmax distribution, which is non-positive, while the text and Table 3 treat υ_i as a positive uncertainty score. There is a missing minus sign, or equivalently an implicit log(1/p) convention, that must be stated. Because υ_i enters every objective through υ_i^γ and is used to identify the least advantaged client, the sign convention is load-bearing.
  5. [Appendix D, Eqs. (26)–(28)] The generalization bound appears to be proved for an objective different from Eq. (8). Eq. (26) defines ZHat{L}^{rβ} as a sum of powered empirical risks, while the dual-norm representation in Eq. (27) yields the rβ-norm (sum_i F_i^{rβ})^{1/(rβ)}. Lemma 2 then bounds L_υ(h) in terms of ||υ||_p ]tilde{L}_{rβ}(h), i.e., the norm, not the sum used in Eq. (26) or in Eq. (8). The authors should state precisely which objective the bound covers and prove the bound for the actual UDJ-FL loss.
minor comments (5)
  1. [§5.2, Rawls' Difference Principle] The text says the Rawls experiments use β→∞, while Table 4 reports β=5; please state explicitly that finite β=5 is an approximation and discuss sensitivity to β.
  2. [§4.4/Algorithm 1] Algorithm 1 treats β as a scalar input, but the desert objective uses a per-client vector β; clarify how the pseudocode and the released implementation handle the vector case.
  3. [§3.3, Eq. (9)] The uncertainty estimate in Eq. (9) depends on model parameters θ, but Algorithm 1 computes υ_i once after S rounds of solo training. The notation θ is reused for the global model; specify at which θ the entropy is evaluated and whether the values are held fixed during federated training.
  4. [§4.1, Eq. (8)] The notation υ_i is introduced as normalized weights υ_i/Σ_j υ_j in Section 4.1, while Eq. (9) defines υ as an average entropy; clarify whether the weights used in Eq. (8) are the normalized values or the raw uncertainties.
  5. [Appendix D, Eq. (28)] The variance term in Eq. (28) uses v_i while the weights elsewhere are υ_i; align the notation.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the UDJ-FL objectives are not defined by the fairness metrics and no load-bearing self-citation is used; the Table 2 reductions for desert and Rawls are algebraically unsupported, which is a correctness issue rather than circularity.

full rationale

I walked the derivation chain from Eq. 5 (Lan et al.'s resource-allocation fairness measure) through Eq. 8 (the UDJ-FL objective) to the four special cases in Table 2 and the fairness metrics in Eqs. 11 and 15. No step is circular under the review's definitions. Eq. 8 is a reweighted, power-transformed empirical-risk objective; the Rawls, egalitarian, and utilitarian metrics are evaluated post hoc on test accuracies, so a high metric value is an empirical outcome rather than a tautology. The desert objective is chosen so that the client weights are monotone in inverse uncertainty, and the desert metric is the correlation between accuracy and uncertainty, but the objective is not literally the metric, and the paper reports the resulting correlations as empirical results. The only load-bearing citations, [19] and [22], are external works, not self-citations, and the solver in Algorithm 1 is borrowed transparently from q-FedAvg [22]. I therefore find no circularity. I do flag a serious non-circular derivation problem: Eq. 8 with r=1, gamma=0 gives sum_i H_i(theta)^{beta} (or, with per-client beta, sum_i H_i^{beta_i}), not the printed desert objective (1/N) sum_i H_i(theta)^{-beta_i} in Eq. 14; and Eq. 8 with r=1+1/beta and gamma=1 gives sum_i upsilon_i H_i^{1+beta}, not Eq. 10's sum_i upsilon_i^{1+beta} H_i^{1+beta}. These mismatches mean the central claim that one objective analytically recovers all four fairness notions is not established by the stated algebra. This is a correctness or omitted-proof concern, not a circularity, so it does not raise the circularity score. The paper's own Section 5.3 limitation, 'we make the assumption that aleatoric uncertainty accurately captures client contribution and advantage,' is likewise a validity caveat, not a circular step.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central claim rests on two pillars: the choice of hyperparameters (beta, r, gamma) that map Eq. 8 to each fairness notion, and the measurement of client quality via aleatoric uncertainty (Eq. 9). The hyperparameters are selected per benchmark after seeing results, so they act as free parameters in the empirical demonstration. The uncertainty-as-contribution assumption is an explicit domain assumption that the authors themselves flag as fragile. No new physical or conceptual entities are introduced.

free parameters (3)
  • beta (loss exponent) = Rawls: 5; Egalitarian: 1; Utilitarian: 0 or 0.1; Desert: beta_i from Eq. 13
    Beta values are chosen after running the experiments to give the best fairness metric on the test sets for each dataset and fairness notion. The paper reports only the winning values in Table 4 and does not derive them from theory or from a held-out model selection procedure.
  • gamma (client weight exponent) = Rawls: 1; Egalitarian: 1; Utilitarian: -1; Desert: 0
    Set by hand for each fairness type (Table 2). The choice is motivated informally in Section 4 but is not derived from the resource-allocation axiomatic framework.
  • r (growth-rate parameter) = Rawls: 1+1/beta; Egalitarian: 1; Utilitarian: 1+1/beta tending to 0; Desert: 1
    Chosen per fairness type in Table 2 with no independent justification; combined with beta it controls the loss exponent in Eq. 8.
assumptions (3)
  • domain assumption Average softmax entropy over a client's local training set is a valid estimator of aleatoric uncertainty and of client data quality, advantage, and contribution.
    Used in Eq. 9 and in all four objectives to define who is least advantaged or who contributes most. The authors acknowledge in Section 5.3 that this may not hold for highly noisy or adversarial data distributions.
  • ad hoc to paper The resource-allocation fairness function of Lan et al. (Eq. 5) can be transformed into a federated learning loss by changing the sign of the exponent and adding per-client weight factors.
    In Section 4.1 the paper asserts this transformation without proof. The sign flip changes the mathematical nature of the objective, and the gamma weighting is introduced arbitrarily, so the resulting Eq. 8 does not formally inherit the fairness properties of Eq. 5.
  • domain assumption The federated setting is horizontal cross-silo with a small number of clients, all participating every round.
    Stated at the start of Section 4. The desert fairness definition relies on equal participation so that contribution cannot be measured by number of rounds, and the uncertainty weights are assumed stable because training is not interrupted.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Achieving Distributive Justice in Federated Learning via Uncertainty Quantification." pith.science (2026). https://pith.science/paper/P6SXAF2I

@misc{pith2026250415924,
  author       = {Pith},
  title        = {Pith review of: Achieving Distributive Justice in Federated Learning via Uncertainty Quantification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P6SXAF2I}},
  note         = {Machine review of arXiv:2504.15924}
}
read the original abstract

Client-level fairness metrics for federated learning are used to ensure that all clients in a federation either: a) have similar final performance on their local data distributions (i.e., client parity), or b) obtain final performance on their local data distributions relative to their contribution to the federated learning process (i.e., contribution fairness). While a handful of works that propose either client-parity or contribution-based fairness metrics ground their definitions and decisions in social theories of equality -- such as distributive justice -- most works arbitrarily choose what notion of fairness to align with which makes it difficult for practitioners to choose which fairness metric aligns best with their fairness ethics. In this work, we propose UDJ-FL (Uncertainty-based Distributive Justice for Federated Learning), a flexible federated learning framework that can achieve multiple distributive justice-based client-level fairness metrics. Namely, by utilizing techniques inspired by fair resource allocation, in conjunction with performing aleatoric uncertainty-based client weighing, our UDJ-FL framework is able to achieve egalitarian, utilitarian, Rawls' difference principle, or desert-based client-level fairness. We empirically show the ability of UDJ-FL to achieve all four defined distributive justice-based client-level fairness metrics in addition to providing fairness equivalent to (or surpassing) other popular fair federated learning works. Further, we provide justification for why aleatoric uncertainty weighing is necessary to the construction of our UDJ-FL framework as well as derive theoretical guarantees for the generalization bounds of UDJ-FL. Our code is publicly available at https://github.com/alycia-noel/UDJ-FL.

Figures

Figures reproduced from arXiv: 2504.15924 by the authors.

Figure 1
Figure 1. Reproduced from [30] and [19]. Rela￾tionship of the four distributive justice theories in terms of the benefit to the most advantaged group and the least advantaged group. The fair￾ness function presented in Eq. 8 can generate any point on the curve depending on the set hyperparameters. we only consider the absolute difference between the results under fed￾erated averaging and results under the chosen fairness metho… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 28 canonical work pages

  1. [1]

    Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mohammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U Rajendra Acharya, et al . 2021. A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information fusion 76 (2021), 243–297

    work page 2021

  2. [2]

    Robert B Ash. 2012. Information theory. Courier Corporation, New York, NY

    work page 2012

  3. [3]

    Flavia Barsotti and Rüya Gökhan Koçer. 2024. MinMax fairness: from Rawlsian Theory of Justice to solution for algorithmic bias. AI & SOCIETY 39, 3 (2024), 961–974

    work page 2024

  4. [4]

    Shrey Bhatt, Aishwarya Gupta, and Piyush Rai. 2024. Federated Learning with Uncertainty via Distilled Predictive Distributions. In Asian Conference on Machine Learning. PMLR, Hanoi, Vietnam, 153–168

    work page 2024

  5. [5]

    Huili Chen, Jie Ding, Eric W Tramel, Shuang Wu, Anit Kumar Sahu, Salman Avestimehr, and Tao Zhang. 2022. Self-aware personalized federated learning. Advances in Neural Information Processing Systems 35 (2022), 20675–20688

    work page 2022

  6. [6]

    Wenda Chu, Chulin Xie, Boxin Wang, Linyi Li, Lang Yin, Han Zhao, and Bo Li. 2022. Focus: Fairness via agent-awareness for federated learning on heterogeneous data. arXiv preprint arXiv:2207.10265 (2022)

    arXiv 2022

  7. [7]

    Kate Donahue and Jon Kleinberg. 2023. Fairness in model-sharing games. In Proceedings of the ACM Web Conference 2023 . 3775–3783

    work page 2023

  8. [8]

    Jiashi Gao, Changwu Huang, Ming Tang, Shin Hwei Tan, Xin Yao, and Xuetao Wei. 2023. EFFL: Egalitarian Fairness in Federated Learning for Mitigating Matthew Effect. arXiv preprint arXiv:2309.16338 (2023)

    work page Pith review arXiv 2023

Show all 40 references
  1. [9]

    Shayan Mohajer Hamidi and En-Hui Yang. 2024. AdaFed: Fair Federated Learning via Adaptive Common Descent Direction. arXiv preprint arXiv:2401.04993 (2024)

    2024 arXiv

  2. [10]

    Zeou Hu, Kiarash Shaloudegi, Guojun Zhang, and Yaoliang Yu. 2020. Fedmgda+: Federated learning meets multi-objective optimization. arXiv preprint arXiv:2006.11489 (2020)

    2020 arXiv

  3. [11]

    Eyke Hüllermeier and Willem Waegeman. 2021. Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods. Machine learning 110, 3 (2021), 457–506

    2021

  4. [12]

    Rajendra K Jain, Dah-Ming W Chiu, William R Hawe, et al. 1984. A quantitative measure of fairness and discrimination. Eastern Research Laboratory, Digital Equipment Corporation, Hudson, MA 21 (1984), 1. Manuscript submitted to ACM 16 Carey and Wu

    1984

  5. [13]

    Simin Javaherian, Sanjeev Panta, Shelby Williams, Md Sirajul Islam, and Li Chen. 2024. FedFairˆ 3: Unlocking Threefold Fairness in Federated Learning. arXiv preprint arXiv:2401.16350 (2024)

    2024 arXiv

  6. [14]

    Li Ju, Tianru Zhang, Salman Toor, and Andreas Hellander. 2024. Accelerating fair federated learning: Adaptive federated adam. IEEE Transactions on Machine Learning in Communications and Networking (2024)

    2024

  7. [15]

    Peter Kairouz, H Brendan McMahan, Brendan Avent, Aurélien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Kallista Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, et al. 2021. Advances and open problems in federated learning. Foundations and trends® in machine learning ...

    2021

  8. [16]

    Selim Kuzucu, Jiaee Cheong, Hatice Gunes, and Sinan Kalkan. 2023. Uncertainty-based fairness measures. arXiv preprint arXiv:2312.11299 (2023)

    2023 arXiv

  9. [17]

    Salem Lahlou, Moksh Jain, Hadi Nekoei, Victor Ion Butoi, Paul Bertin, Jarrid Rector-Brooks, Maksym Korablyov, and Yoshua Bengio. 2021. Deup: Direct epistemic uncertainty prediction. arXiv preprint arXiv:2102.08501 (2021)

    2021 arXiv

  10. [18]

    Julian Lamont and Christi Favor. 2017. Distributive Justice. In The Stanford Encyclopedia of Philosophy (Winter 2017 ed.), Edward N. Zalta (Ed.). Metaphysics Research Lab, Stanford University

    2017

  11. [19]

    Tian Lan, David Kao, Mung Chiang, and Ashutosh Sabharwal. 2010. An axiomatic theory of fairness in network resource allocation . IEEE

    2010

  12. [20]

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. 1998. Gradient-based learning applied to document recognition. Proc. IEEE 86, 11 (1998), 2278–2324

    1998

  13. [21]

    Tian Li, Ahmad Beirami, Maziar Sanjabi, and Virginia Smith. 2020. Tilted empirical risk minimization. arXiv preprint arXiv:2007.01162 (2020)

    2020 arXiv

  14. [22]

    Tian Li, Maziar Sanjabi, Ahmad Beirami, and Virginia Smith. 2019. Fair resource allocation in federated learning. arXiv preprint arXiv:1905.10497 (2019)

    2019 arXiv

  15. [23]

    Lingjuan Lyu, Xinyi Xu, Qian Wang, and Han Yu. 2020. Collaborative fairness in federated learning. Federated Learning: Privacy and Incentive (2020), 189–204

    2020

  16. [24]

    Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics. PMLR, 1273–1282

    2017

  17. [25]

    Jeonghoon Mo and Jean Walrand. 2000. Fair end-to-end window-based congestion control. IEEE/ACM Transactions on networking 8, 5 (2000), 556–567

    2000

  18. [26]

    Mehryar Mohri, Gary Sivek, and Ananda Theertha Suresh. 2019. Agnostic federated learning. In International conference on machine learning . PMLR, 4615–4625

    2019

  19. [27]

    Hamid Mozaffari and Amir Houmansadr. 2022. E2FL: Equal and equitable federated learning. arXiv preprint arXiv:2205.10454 (2022)

    2022 arXiv

  20. [28]

    Jishnu Mukhoti, Andreas Kirsch, Joost van Amersfoort, Philip HS Torr, and Yarin Gal. 2023. Deep deterministic uncertainty: A new simple baseline. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 24384–24394

    2023

  21. [29]

    Dinh C Nguyen, Ming Ding, Pubudu N Pathirana, Aruna Seneviratne, Jun Li, and H Vincent Poor. 2021. Federated learning for internet of things: A comprehensive survey. IEEE Communications Surveys & Tutorials 23, 3 (2021), 1622–1658

    2021

  22. [30]

    John Rawls. 1971. Atheory of justice. Cambridge (Mass.) (1971)

    1971

  23. [31]

    Bhaskar Ray Chaudhury, Linyi Li, Mintong Kang, Bo Li, and Ruta Mehta. 2022. Fairness in federated learning via core-stability. Advances in neural information processing systems 35 (2022), 5738–5750

    2022

  24. [32]

    Cong Su, Guoxian Yu, Jun Wang, Hui Li, Qingzhong Li, and Han Yu. 2024. Multi-Dimensional Fair Federated Learning. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 38. 15083–15090

    2024

  25. [33]

    Anique Tahir, Lu Cheng, and Huan Liu. 2023. Fairness through aleatoric uncertainty. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management . PMLR, 2372–2381

    2023

  26. [34]

    Temel, G

    D. Temel, G. Kwon, M. Prabhushankar, and G. AlRegib. 2017. CURE-TSR: Challenging unreal and real environments for traffic sign recognition. In Neural Information Processing Systems (NIPS) Workshop on Machine Learning for Intelligent Transportation Systems (MLITS)

    2017

  27. [35]

    Lin Wang, Zhichao Wang, and Xiaoying Tang. 2023. Fedeba+: Towards fair and effective federated learning via entropy-based model. arXiv preprint arXiv:2301.12407 (2023)

    2023 arXiv

  28. [36]

    Zheng Wang, Xiaoliang Fan, Jianzhong Qi, Chenglu Wen, Cheng Wang, and Rongshan Yu. 2021. Federated learning with fair averaging. arXiv preprint arXiv:2104.14937 (2021)

    2021 arXiv

  29. [37]

    Huang Wei, Tianrui Li, Dexian Wang, Shengdong Du, and Junbo Zhang. 2020. Fairness and accuracy in federated learning. arXiv preprint (2020)

    2020

  30. [38]

    Lisa Wimmer, Yusuf Sale, Paul Hofman, Bernd Bischl, and Eyke Hüllermeier. 2023. Quantifying aleatoric and epistemic uncertainty in machine learning: Are conditional entropy and mutual information appropriate measures?. In Uncertainty in Artificial Intelligence. PMLR, 2282–2292

    2023

  31. [39]

    Jie Xu, Benjamin S Glicksberg, Chang Su, Peter Walker, Jiang Bian, and Fei Wang. 2021. Federated learning for healthcare informatics. Journal of healthcare informatics research 5 (2021), 1–19

    2021

  32. [40]

    you get what you deserve

    Guojun Zhang, Saber Malekmohammadi, Xi Chen, and Yaoliang Yu. 2022. Proportional fairness in federated learning.arXiv preprint arXiv:2202.01666 (2022). Manuscript submitted to ACM Achieving Distributive Justice in Federated Learning via Uncertainty Quantification 17 A Motivati...

    2022 arXiv

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.