Pith. sign in

REVIEW 3 major objections 4 minor 146 references

Personalized Federated Learning via Variance-Aware Nonparametric Empirical Bayes

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Personalized federated learning can be cast as a variance-aware empirical Bayes problem: with a generalized Tweedie formula, the excess risk over the oracle Bayes estimator is near $\tilde{O}((nK)^{-1})$ up to polylog factors.

desk verdict A serious NPEB paper with a real contribution, but Theorem 4.2's hypothesis and proof are inconsistent and the main FL claim is not established until that is fixed. read the letter →

arxiv 2608.09074 v1 pith:KKZUQFRJ submitted 2026-08-10 stat.ML cs.LGstat.ME

Jae Ho Chang , Arnab Auddy , Subhadeep Paul This is my paper · ORCID
classification stat.MLcs.LGstat.ME MSC 62C1262G0562F15
keywords federatedlearningpersonalizationnonparametricempiricalBayesmaximumlikelihoodTweedie'sformulaheteroskedasticityoracledenoisinginequalitydeepneuralnetworks
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 tries to establish that personalized federated learning across heterogeneous clients is a compound decision problem: each client's asymptotically normal local estimator is a noisy observation of a client-specific parameter, all parameters are drawn from one unknown shared prior, and the server only has to learn that prior from the transmitted summaries. Its procedure, VANEB, handles the realistic case where the local noise covariance depends on the unknown parameter, which standard nonparametric empirical Bayes cannot. The main quantitative claim is an oracle denoising inequality: after substituting the estimated prior into a generalized Tweedie formula, the average squared error across clients is within roughly $\tilde{O}((nK)^{-1})$ (up to polylogarithmic factors in $K$) of the infeasible oracle posterior mean, in discrete, compact-support, and Gaussian-mixture settings. If this bound is right, each client's personalized estimate can be almost as good as the posterior mean under the true population of client parameters, and the entire federation gain comes from learning that population. The paper also gives neural-network extensions, VANEB-Head and VANEB-FT, and reports that the fine-tuned variant has the highest reported accuracy under balanced evaluation on MNIST and CIFAR-10.

What carries the argument

The load-bearing object is the variance-aware NPMLE for the unknown prior $G_0$, computed by a pseudo-EM algorithm that alternates between fixed-covariance ridgeline updates and re-evaluating the covariance function at the updated atoms. The identity that carries the argument is the generalized Tweedie formula, which expresses the oracle posterior mean through variance-weighted marginal densities; the coordinate weights $\sigma_{k,i}(\theta)^2$ make shrinkage automatically stronger where the local estimate is least reliable. The oracle inequality then routes through the Hellinger convergence of both the ordinary and variance-weighted marginals, with the condition number $\tau=\bar{s}/s$ controlling the transfer between them. In the deep-learning extension, the same operation is applied only to the last-layer heads, using a diagonal Fisher-information estimate of the head covariance.

What would settle it

Simulate the paper's quadratic-variance scenario with the covariance function known exactly, as Assumption 1 requires, and measure the average squared excess risk against the true oracle posterior mean at fixed $n$ while $K$ grows; Theorem 4.2 predicts decay like $(\log K)^{3\vee(d/2-1)}\varepsilon_M^2/n$ with $\varepsilon_M^2\approx (\log^{d/2+1}K)/K$ for compact support. If the empirical excess risk fails to decrease with $K$ once the prior is nontrivial, the oracle denoising inequality is refuted.

Watch

Extended reading notes

Core claim

The paper's central discovery is a generalized Tweedie formula for heteroskedastic Gaussian observations whose covariance matrix is a known function of the unknown parameter. Because the covariance cannot be pulled out of the integral as in the classical formula, the paper defines variance-weighted marginal densities $\tilde{f}_{k,G,i}(\hat\theta_k)=\int \sigma_{k,i}(\theta)^2\varphi^{(k)}(\hat\theta_k;\theta)\,dG(\theta)$ and shows that the oracle Bayes rule is $\theta_k^o=\hat\theta_k/\sqrt{n_k}+q_{k,G_0}(\hat\theta_k)/(\sqrt{n_k}f_{k,G_0}(\hat\theta_k))$, where $q_{k,G}$ is the integral of the covariance-weighted likelihood score. Estimating $G_0$ by the nonparametric maximum likelihood estimator and plugging it into this formula gives the VANEB estimator. The proof transfers Hellinger accuracy of the NPMLE to the variance-weighted densities through the condition number $\tau=\bar{s}/s$, then converts that into the oracle denoising inequality of Theorem 4.2, whose excess risk is $(\log K)^{3\vee(d/2-1)}\varepsilon_M^2(S_\bullet,G_0)/n$; in the discrete, compact-support, and Gaussian-mixture cases this is $\tilde{O}((nK)^{-1})$.

Load-bearing premise

The load-bearing premise is the working model that each client's rescaled local estimator is exactly Gaussian around the true parameter with a known, diagonal covariance function of that parameter; in practice that covariance must be estimated from local data, and the paper explicitly sets this added uncertainty aside.

Editorial extensions

If this is right

  • Under the paper's Assumption 1, VANEB's average squared excess risk relative to the oracle posterior mean is $O((\log K)^{3\vee(d/2-1)}\varepsilon_M^2/n)$, so more clients improve personalization even when every local sample size is fixed.
  • For discrete, compactly supported, and Gaussian-mixture priors the bound becomes $\tilde O((nK)^{-1})$, so the federation gain enters entirely through learning the shared prior.
  • Local M-estimators used without pooling have excess risk that cannot fall below the oracle baseline's $1/n$-scale floor as $K$ grows; in VANEB's bound only the prior-estimation term $\varepsilon_M^2$ carries the $K$-dependence.
  • Any approximate NPMLE whose log-likelihood gap is within the rate $\varepsilon_M^2$ inherits the same Hellinger accuracy, which justifies the discretized support-constrained solver used in computation.
  • In the reported balanced evaluations on MNIST and CIFAR-10, VANEB-FT records the highest accuracy among the compared personalized methods, indicating that the shrinkage prevents local adaptation from discarding globally informative classes.

Reading between the lines

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

  • The paper sets aside the uncertainty from estimating the covariance function, so a natural follow-up is a bound with an extra covariance-estimation error term; if that term does not vanish, the advertised $(nK)^{-1}$ rate would degrade.
  • The diagonal-covariance assumption is violated by the paper's own full-covariance simulations and by the DNN head estimator, so the empirical success there suggests the variance-aware mechanism is more robust than the proved regime, not that the theory covers it.
  • A concrete testable extension is to use VANEB as a one-shot server-side post-processor for any asymptotically normal local summary, including differentially private summaries, and to measure how much privacy noise the oracle inequality tolerates.
  • The balanced-evaluation pattern implies that empirical-Bayes shrinkage acts as a regularizer that keeps local heads from collapsing onto the client's observed label mixture; a direct ablation varying the number of NPMLE atoms would show how much of that coverage comes from prior-learning rather than simple averaging.
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.

Referee Report

3 major / 4 minor

Summary. The paper proposes VANEB, a nonparametric empirical Bayes framework for personalized federated learning that treats local M-estimators as noisy observations with a parameter-dependent Gaussian covariance. The authors derive a generalized Tweedie formula for diagonal heteroskedastic covariances, formulate an NPMLE for the shared prior, propose a pseudo-EM algorithm with fixed-point guarantees, and prove non-asymptotic Hellinger rates and an oracle denoising inequality. They then introduce DNN-based extensions, VANEB-Head and VANEB-FT, that personalize the final layer under approximate diagonal Fisher-information covariances, and report experiments on simulations, MNIST, and CIFAR-10. The theoretical results are stated under Assumption 1, which requires the covariance functions to be known and diagonal.

Significance. If the theoretical results are correct, the paper addresses a genuine gap in NPEB methodology: prior work assumes known, parameter-independent covariances, while M-estimators in federated settings naturally produce parameter-dependent, unknown covariance matrices. The generalized Tweedie derivation and the variance-weighted marginal densities are novel and clearly presented, and the oracle denoising inequality would be a strong statement about the benefit of pooling local estimates. The empirical results, especially the balanced-evaluation gains of VANEB-FT, are encouraging. However, the significance is currently conditional: the main theorem's hypothesis appears to be stated with an exponent inconsistent with the proof, and the theory explicitly ignores the covariance-estimation error that is present in both the simulations and the deep-learning applications. These issues prevent the paper, in its present form, from establishing the advertised O((nK)^{-1}) federation gain in the regimes it emphasizes.

major comments (3)
  1. [Section 4.2 and Appendix A.4.3] Theorem 4.2 states its oracle denoising inequality under the hypothesis n ≳ K M^{d-2} log^{d/2+1} K. Since M ≳ √(log K), this is equivalent to n ≳ K log^d K. However, in the proof's bound on E(Q_2^2), the text requires only n ≳ K M^{2-d} log^{-d/2-1} K, which for d>2 is a much weaker condition (for d=3, roughly K/log^3 K rather than K log^3 K). These two conditions are inconsistent. As written, the theorem's central claim of excess risk of order (1/(nK)) up to polylog factors is not established in the fixed-total-sample FL regime reported in Tables 1-2 and Figure 5, where n ≈ 50,000/K and the theorem's hypothesis fails for large K. The authors should either correct the theorem statement or supply the additional proof steps needed to justify the stated condition.
  2. [Assumption 1 and Section 2] Assumption 1 requires the covariance function Σ_k(·) to be known and diagonal, and the working model (1) treats the local estimator as exactly Gaussian with this covariance. The paper explicitly states in Section 2 that 'We ignore the added uncertainty due to this here,' referring to clients reporting estimated covariance functions. This is load-bearing because the theory's Hellinger rates and oracle inequality do not cover the estimated-covariance case, yet the logistic and Poisson simulations use full, data-dependent Fisher information matrices and the DNN experiments use the estimated diagonal covariance in (20). The paper therefore does not currently provide a theoretical guarantee for the actual procedure being proposed for federated learning. The authors should either add an analysis of covariance-estimation error or explicitly restrict the theoretical claims to the known-covariance setting and present the practical method as a heuristic.
  3. [Section 4.2, Gaussian mixture case] In the special-case proof for Gaussian mixture priors, the text argues that the term involving the tail mass 'becomes negligible as K is assumed to diverge' without giving a quantitative rate. Since this special case is one of the headline Õ((nK)^{-1}) claims, the proof should explicitly track the dependence on the mixture parameters A*, γ, the radius r, and K, and show how Vol(S_r) arises in the stated bound. As written, the GMM case is not fully established.
minor comments (4)
  1. [Section 2] There is a typo in the sentence defining the marginal law: 'F zk(z)and and define' should read 'F_{z_k}(z) and define'.
  2. [Section 2.2 and Algorithm 1] The notation θ̂_k is reused for both the original local estimator and its rescaled version √(n_k) θ̂_k. This creates confusion in the likelihood expressions and in Algorithm 1; distinct symbols should be used.
  3. [Section 3.1.3] The text says that 'contraction is controlled by Brouwer's fixed-point theorem,' but Brouwer's theorem gives existence of a fixed point, while contraction gives uniqueness and linear rate; the wording should distinguish these two ingredients.
  4. [Figures 2-4] The axis label 'Log x' in the K-sweep panels is unclear; it appears to indicate a logarithmic horizontal axis, not a quantity named x. Please relabel for clarity.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the generalized Tweedie identity is derived by direct differentiation, the NPMLE rates rest on external benchmarks, and the oracle inequality is a genuine excess-risk bound; the sole self-citation is a non-load-bearing literature mention.

full rationale

The derivation chain is self-contained on the points that matter. The generalized Tweedie formula in Section 2.2 is obtained from the definition q_{k,G}(θ̂_k) = ∫ Σ_k(θ)∇φ^{(k)}(θ̂_k;θ) dG(θ), and the paper then computes q_{k,G}(θ̂_k) = ∫ φ^{(k)}(θ̂_k;θ)(√n_k θ − θ̂_k) dG(θ) by direct coordinate differentiation of the diagonal Gaussian kernel; the posterior-mean identity √n_k E(θ_k | θ̂_k) = θ̂_k + q/f is therefore an algebraic consequence of Assumption 1's Gaussian likelihood, not a fitted input renamed as a prediction. The NPMLE Hellinger rates in Theorem 4.1 and Corollary 4.1.1 apply the entropy and moment-matching framework of Saha and Guntuboyina (2020) and Soloff et al. (2025), which are external benchmarks independent of the present authors, and the log-likelihood-gap condition (11) is stated as an explicit optimization-quality hypothesis rather than assumed away. Theorem 4.2 bounds the excess risk of θ̂^v relative to the infeasible oracle posterior mean θ^o under G0; the oracle rule is not used to fit Ĝ, so the bound is a genuine comparison against an unobserved target rather than a tautology. The only self-citation is the literature-review mention of [Auddy et al., 2026] as one of several FL references; it is not invoked to justify the model, the NPMLE theory, or the oracle inequality. The paper's own limitations—Section 2's 'We ignore the added uncertainty due to this here' about estimated covariance functions and Section 7's 'our guarantees are restricted to the diagonal covariance approximation'—are acknowledged coverage restrictions, not circular inputs. The mismatch between Theorem 4.2's stated hypothesis n ≳ K M^{d−2} log^{d/2+1} K and the proof's 'provided that n ≳ K M^{2−d} log^{−d/2−1} K' in A.4.3 is a correctness or consistency risk for the claimed FL regime (fixed total sample, large K), not an instance of a result reducing to its own definition or to a self-citation.

Assumptions & free parameters 4 free parameters · 4 assumptions · 1 invented entities

The central claim rests on the Gaussian working model and the known-diagonal-covariance assumption. The method introduces no new physical entities, but it does define a variance-weighted marginal density used in the generalized Tweedie formula. The free parameters are experimental hyperparameters; the main unmodeled cost is the ignored covariance estimation error.

free parameters (4)
  • DNN covariance regularizer λ = not reported
    In equation (20), the diagonal covariance estimate uses (1/n_k Σ g g^T + λI)^{-1}; the value of λ is not given in the experiments, and the results may depend on it.
  • Number of atoms m for VANEB = 10
    The VANEB procedures use 10 atoms (Section 6). This is chosen by hand; no sensitivity analysis is provided.
  • EB frequency b = 5
    VANEB-Head applies the EB update every five communication rounds; the choice is arbitrary.
  • s_min, s_max clipping in quadratic variance simulation = 0.01, 100
    Scenario (i) clips the quadratic variance to [0.01,100] for numerical stability; this is a hand-set constant.
assumptions (4)
  • domain assumption Local M-estimators are asymptotically normal with covariance Σ_k(θ_k)/n_k (working model (1))
    Section 2 adopts the Gaussian working model from asymptotic normality of M-estimators; the entire analysis is conditional on this approximation.
  • domain assumption Assumption 1: Σ_k known, diagonal, uniformly elliptic, Lipschitz with C1 log-variance
    Stated before Theorem 4.1; all theoretical results require it.
  • ad hoc to paper Covariance estimation error is negligible (ignored)
    Section 2: 'We ignore the added uncertainty due to this here' when clients report estimated covariance functions. This is load-bearing because real FL clients do not know Σ_k exactly.
  • ad hoc to paper Approximate Gaussianity of local DNN parameters
    Section 6 acknowledges the asymptotic normality assumption does not hold for non-convex neural networks, yet the method is applied heuristically.
invented entities (1)
  • Variance-weighted marginal density f̃_{k,G,i}
    purpose: Used to express the generalized Tweedie formula and transfer Hellinger bounds between f_{k,G} and the score q_{k,G}
    This is a new functional of the prior, defined in Section 2.2; it has no falsifiable handle outside the paper's own theorems.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Personalized Federated Learning via Variance-Aware Nonparametric Empirical Bayes." pith.science (2026). https://pith.science/paper/KKZUQFRJ

@misc{pith2026260809074,
  author       = {Pith},
  title        = {Pith review of: Personalized Federated Learning via Variance-Aware Nonparametric Empirical Bayes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KKZUQFRJ}},
  note         = {Machine review of arXiv:2608.09074}
}
read the original abstract

We develop a new approach to Personalized Federated Learning across heterogeneous clients using Nonparametric Empirical Bayes (NPEB). Leveraging the asymptotic normality of local parameter estimates obtained from Empirical Risk Minimization or M-estimation, our method formulates these estimates as noisy observations to estimate an unknown shared prior via Nonparametric Maximum Likelihood. A key challenge in applying NPEB in this setting is that existing approaches assume known fixed variances, which is not true in practice. To address this, we introduce a Variance-Aware Nonparametric Empirical Bayes (VANEB) framework that leverages the parameter-dependent asymptotic variance of local M-estimators. A key technical contribution is a generalized Tweedie's formula for this heteroskedastic setting. We then establish non-asymptotic error rates for density estimation in the average squared Hellinger distance and derive an oracle denoising inequality that provides error bounds for our estimator. While our theoretical guarantees are rooted in the asymptotic regime of M-estimators, we empirically explore heuristic extensions of VANEB to modern federated learning settings involving Deep Neural Networks (DNNs). For DNNs, we propose VANEB-head and VANEB-FT, which personalize the last fully connected layer via an NPEB step using an approximate diagonal variance estimator. We show that our method has strong performance on popular vision datasets MNIST and CIFAR-10, using a convolutional neural network architecture.

Figures

Figures reproduced from arXiv: 2608.09074 by the authors.

Figure 1
Figure 1. Support of G0 24 [PITH_FULL_IMAGE:figures/full_fig_p024_1.png] view at source ↗
Figure 2
Figure 2. Quadratic variance scenario: RMSE comparison ( [PITH_FULL_IMAGE:figures/full_fig_p027_2.png] view at source ↗
Figure 3
Figure 3. Multiclass logistic scenario: RMSE comparison. FedAvg omitted from part (a) [PITH_FULL_IMAGE:figures/full_fig_p028_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Poisson regression scenario: RMSE comparison. [PITH_FULL_IMAGE:figures/full_fig_p029_4.png]
Figure 5
Figure 5. Figure 5: Final personalized accuracy on MNIST as the number of clients increases while [PITH_FULL_IMAGE:figures/full_fig_p036_5.png]
Figure 6
Figure 6. Figure 6: Estimated VANEB-FT priors on MNIST with α = 0.1 and a fixed total training sample size of 50,000. The top and bottom rows within each subfigure show the coordi￾natewise and direct covariance constructions, respectively. 38 [PITH_FULL_IMAGE:figures/full_fig_p038_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

146 extracted references · 33 canonical work pages

  1. [1]

    Journal of Machine Learning Research , volume=

    Minimax Estimation for Personalized Federated Learning: An Alternative between FedAvg and Local Training? , author=. Journal of Machine Learning Research , volume=

  2. [2]

    arXiv preprint arXiv:2003.00295 , year=

    Adaptive federated optimization , author=. arXiv preprint arXiv:2003.00295 , year=

    arXiv 2003

  3. [3]

    , author=

    Lora: Low-rank adaptation of large language models. , author=. Iclr , volume=

  4. [4]

    Proceedings of the 2024 conference on empirical methods in natural language processing , pages=

    Heterogeneous lora for federated fine-tuning of on-device foundation models , author=. Proceedings of the 2024 conference on empirical methods in natural language processing , pages=

    2024

  5. [5]

    arXiv preprint arXiv:2106.06042 , year=

    Fedbabu: Towards enhanced representation for federated image classification , author=. arXiv preprint arXiv:2106.06042 , year=

    arXiv

  6. [6]

    arXiv preprint arXiv:2001.01523 , year=

    Think locally, act globally: Federated learning with local and global representations , author=. arXiv preprint arXiv:2001.01523 , year=

    arXiv 2001

  7. [7]

    arXiv preprint arXiv:1912.00818 , year=

    Federated learning with personalization layers , author=. arXiv preprint arXiv:1912.00818 , year=

    arXiv 1912

  8. [8]

    International conference on machine learning , pages=

    Ditto: Fair and robust federated learning through personalization , author=. International conference on machine learning , pages=. 2021 , organization=

    2021

Show all 146 references
  1. [9]

    arXiv preprint arXiv:2308.06522 , year=

    Slora: Federated parameter efficient fine-tuning of language models , author=. arXiv preprint arXiv:2308.06522 , year=

    arXiv

  2. [10]

    Advances in neural information processing systems , volume=

    Personalized federated learning with moreau envelopes , author=. Advances in neural information processing systems , volume=

  3. [11]

    arXiv preprint arXiv:2409.15723 , year=

    Federated large language models: Current progress and future directions , author=. arXiv preprint arXiv:2409.15723 , year=

    arXiv

  4. [12]

    arXiv preprint arXiv:2306.15546 , year=

    When foundation model meets federated learning: Motivations, challenges, and future directions , author=. arXiv preprint arXiv:2306.15546 , year=

    arXiv

  5. [13]

    arXiv preprint arXiv:2002.07948 , year=

    Personalized federated learning: A meta-learning approach , author=. arXiv preprint arXiv:2002.07948 , year=

    2002 arXiv

  6. [14]

    arXiv preprint arXiv:1909.12488 , year=

    Improving federated learning personalization via model agnostic meta learning , author=. arXiv preprint arXiv:1909.12488 , year=

    1909 arXiv

  7. [15]

    IEEE transactions on neural networks and learning systems , volume=

    Towards personalized federated learning , author=. IEEE transactions on neural networks and learning systems , volume=. 2022 , publisher=

    2022

  8. [16]

    Advances in Neural Information Processing Systems , volume=

    Multiply robust federated estimation of targeted average treatment effects , author=. Advances in Neural Information Processing Systems , volume=

  9. [17]

    International Conference on Artificial Intelligence and Statistics , pages=

    Fedfisher: Leveraging fisher information for one-shot federated learning , author=. International Conference on Artificial Intelligence and Statistics , pages=. 2024 , organization=

    2024

  10. [18]

    Statistics in Medicine , volume=

    Federated causal inference in heterogeneous observational data , author=. Statistics in Medicine , volume=. 2023 , publisher=

    2023

  11. [19]

    Proceedings of the 2019 ACM SIGSAC Conference on Cloud Computing Security Workshop , pages=

    PrivFL: Practical privacy-preserving federated regressions on high-dimensional data over mobile networks , author=. Proceedings of the 2019 ACM SIGSAC Conference on Cloud Computing Security Workshop , pages=

    2019

  12. [20]

    International Conference on Artificial Intelligence and Statistics , pages=

    Federated asymptotics: a model to compare federated learning algorithms , author=. International Conference on Artificial Intelligence and Statistics , pages=. 2023 , organization=

    2023

  13. [21]

    Foundations and trends

    Advances and open problems in federated learning , author=. Foundations and trends. 2021 , publisher=

    2021

  14. [22]

    arXiv preprint arXiv:2403.11343 , year=

    Federated Transfer Learning with Differential Privacy , author=. arXiv preprint arXiv:2403.11343 , year=

    arXiv

  15. [23]

    Scientific reports , volume=

    Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data , author=. Scientific reports , volume=. 2020 , publisher=

    2020

  16. [24]

    Journal of healthcare informatics research , volume=

    Federated learning for healthcare informatics , author=. Journal of healthcare informatics research , volume=. 2021 , publisher=

    2021

  17. [25]

    IEEE Communications Surveys & Tutorials , volume=

    Federated learning for internet of things: A comprehensive survey , author=. IEEE Communications Surveys & Tutorials , volume=. 2021 , publisher=

    2021

  18. [26]

    NPJ digital medicine , volume=

    The future of digital health with federated learning , author=. NPJ digital medicine , volume=. 2020 , publisher=

    2020

  19. [27]

    Proceedings of Machine learning and systems , volume=

    Federated optimization in heterogeneous networks , author=. Proceedings of Machine learning and systems , volume=

  20. [28]

    arXiv preprint arXiv:2003.13461 , year=

    Adaptive personalized federated learning , author=. arXiv preprint arXiv:2003.13461 , year=

    2003 arXiv

  21. [29]

    arXiv preprint arXiv:2002.10619 , year=

    Three approaches for personalization with applications to federated learning , author=. arXiv preprint arXiv:2002.10619 , year=

    2002 arXiv

  22. [30]

    Advances in neural information processing systems , volume=

    Federated multi-task learning , author=. Advances in neural information processing systems , volume=

  23. [31]

    IEEE transactions on neural networks and learning systems , volume=

    Clustered federated learning: Model-agnostic distributed multitask optimization under privacy constraints , author=. IEEE transactions on neural networks and learning systems , volume=. 2020 , publisher=

    2020

  24. [32]

    Advances in neural information processing systems , volume=

    An efficient framework for clustered federated learning , author=. Advances in neural information processing systems , volume=

  25. [33]

    Advances in Neural Information Processing Systems , volume=

    A bayesian approach for personalized federated learning in heterogeneous settings , author=. Advances in Neural Information Processing Systems , volume=

  26. [34]

    Statistical science: a review journal of the Institute of Mathematical Statistics , volume=

    Two modeling strategies for empirical Bayes estimation , author=. Statistical science: a review journal of the Institute of Mathematical Statistics , volume=

  27. [35]

    Proceedings of ICML workshop on unsupervised and transfer learning , pages=

    Deep learning of representations for unsupervised and transfer learning , author=. Proceedings of ICML workshop on unsupervised and transfer learning , pages=. 2012 , organization=

    2012

  28. [36]

    Advances in neural information processing systems , volume=

    On the theory of transfer learning: The importance of task diversity , author=. Advances in neural information processing systems , volume=

  29. [37]

    arXiv preprint arXiv:1910.01108 , year=

    DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter , author=. arXiv preprint arXiv:1910.01108 , year=

    1910 arXiv

  30. [38]

    OpenAI blog , volume=

    Language models are unsupervised multitask learners , author=. OpenAI blog , volume=

  31. [39]

    Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability , volume=

    Estimation with quadratic loss , author=. Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability , volume=. 1961 , organization=

    1961

  32. [40]

    Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability , volume=

    Inadmissibility of the usual estimator for the mean of a multivariate normal distribution , author=. Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability , volume=. 1956 , publisher=

    1956

  33. [41]

    2012 , publisher=

    Large-Scale Inference: Empirical Bayes Methods for Estimation, Testing, and Prediction , author=. 2012 , publisher=

    2012

  34. [42]

    arXiv preprint arXiv:2009.09459 , year=

    Advantage of multitask learning for federated personalization , author=. arXiv preprint arXiv:2009.09459 , year=

    2009 arXiv

  35. [43]

    International conference on machine learning , pages=

    Exploiting shared representations for personalized federated learning , author=. International conference on machine learning , pages=. 2021 , organization=

    2021

  36. [44]

    arXiv preprint arXiv:2605.18656 , year=

    Statistical Limits and Efficient Algorithms for Differentially Private Federated Learning , author=. arXiv preprint arXiv:2605.18656 , year=

    arXiv

  37. [45]

    The Annals of Statistics , volume=

    The cost of privacy: Optimal rates of convergence for parameter estimation with differential privacy , author=. The Annals of Statistics , volume=. 2021 , publisher=

    2021

  38. [46]

    Electronic Journal of Statistics , volume=

    Empirical Bayes inference in sparse high-dimensional generalized linear models , author=. Electronic Journal of Statistics , volume=. 2024 , publisher=

    2024

  39. [47]

    Advances in Neural Information Processing Systems , volume=

    Nearly optimal private lasso , author=. Advances in Neural Information Processing Systems , volume=

  40. [48]

    arXiv preprint arXiv:2406.06755 , year=

    Optimal Federated Learning for Nonparametric Regression with Heterogeneous Distributed Differential Privacy Constraints , author=. arXiv preprint arXiv:2406.06755 , year=

    arXiv

  41. [49]

    arXiv preprint arXiv:2406.20088 , year=

    Minimax And Adaptive Transfer Learning for Nonparametric Classification under Distributed Differential Privacy Constraints , author=. arXiv preprint arXiv:2406.20088 , year=

    arXiv

  42. [50]

    arXiv preprint arXiv:2401.03820 , year=

    Optimal Differentially Private PCA and Estimation for Spiked Covariance Matrices , author=. arXiv preprint arXiv:2401.03820 , year=

    arXiv

  43. [51]

    International Conference on Learning Representations (ICLR), 2023

    A statistical framework for personalized federated learning and estimation: Theory, algorithms, and privacy , author=. International Conference on Learning Representations (ICLR), 2023. , year=

    2023

  44. [52]

    International Conference on Machine Learning , pages=

    On sparse linear regression in the local differential privacy model , author=. International Conference on Machine Learning , pages=. 2019 , organization=

    2019

  45. [53]

    The Annals of Statistics , volume=

    The right complexity measure in locally private estimation: It is not the fisher information , author=. The Annals of Statistics , volume=. 2024 , publisher=

    2024

  46. [54]

    2013 IEEE 54th annual symposium on foundations of computer science , pages=

    Local privacy and statistical minimax rates , author=. 2013 IEEE 54th annual symposium on foundations of computer science , pages=. 2013 , organization=

    2013

  47. [55]

    IEEE Internet of Things Journal , volume=

    Personalized federated learning with differential privacy , author=. IEEE Internet of Things Journal , volume=. 2020 , publisher=

    2020

  48. [56]

    IEEE transactions on information forensics and security , volume=

    Federated learning with differential privacy: Algorithms and performance analysis , author=. IEEE transactions on information forensics and security , volume=. 2020 , publisher=

    2020

  49. [57]

    International colloquium on automata, languages, and programming , pages=

    Differential privacy , author=. International colloquium on automata, languages, and programming , pages=. 2006 , organization=

    2006

  50. [58]

    International conference on theory and applications of models of computation , pages=

    Differential privacy: A survey of results , author=. International conference on theory and applications of models of computation , pages=. 2008 , organization=

    2008

  51. [59]

    Journal of the American Statistical Association , volume=

    A statistical framework for differential privacy , author=. Journal of the American Statistical Association , volume=. 2010 , publisher=

    2010

  52. [60]

    Management Science , volume=

    Predicting with proxies: Transfer learning in high dimension , author=. Management Science , volume=. 2021 , publisher=

    2021

  53. [61]

    Journal of the American Statistical Association , volume=

    Transfer learning in large-scale gaussian graphical models with false discovery rate control , author=. Journal of the American Statistical Association , volume=. 2023 , publisher=

    2023

  54. [62]

    Journal of the American Statistical Association , pages=

    Transfer learning in large-scale gaussian graphical models with false discovery rate control , author=. Journal of the American Statistical Association , pages=. 2022 , publisher=

    2022

  55. [63]

    Journal of the American Statistical Association , pages=

    Transfer learning under high-dimensional generalized linear models , author=. Journal of the American Statistical Association , pages=. 2022 , publisher=

    2022

  56. [64]

    Journal of the American Statistical Association , pages=

    Estimation and inference for high-dimensional generalized linear models with knowledge transfer , author=. Journal of the American Statistical Association , pages=. 2023 , publisher=

    2023

  57. [65]

    2018 , publisher=

    Double/debiased machine learning for treatment and structural parameters , author=. 2018 , publisher=

    2018

  58. [66]

    The Journal of Machine Learning Research , volume=

    Confidence intervals and hypothesis testing for high-dimensional regression , author=. The Journal of Machine Learning Research , volume=. 2014 , publisher=

    2014

  59. [67]

    On asymptotically optimal confidence regions and tests for high-dimensional models , author=

  60. [68]

    Advances in Neural Information Processing Systems , volume=

    Transfer learning on heterogeneous feature spaces for treatment effects estimation , author=. Advances in Neural Information Processing Systems , volume=

  61. [69]

    arXiv preprint arXiv:2305.09126 , year=

    Transfer causal learning: Causal effect estimation with knowledge transfer , author=. arXiv preprint arXiv:2305.09126 , year=

    arXiv

  62. [70]

    Journal of the American Medical Informatics Association , volume=

    A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions , author=. Journal of the American Medical Informatics Association , volume=. 2014 , publisher=

    2014

  63. [71]

    Statistics in Medicine , volume=

    Bayesian federated inference for estimating statistical models based on non-shared multicenter data sets , author=. Statistics in Medicine , volume=. 2024 , publisher=

    2024

  64. [72]

    International Conference on Machine Learning , pages=

    RadialGAN: Leveraging multiple datasets to improve target-specific predictive models using Generative Adversarial Networks , author=. International Conference on Machine Learning , pages=. 2018 , organization=

    2018

  65. [73]

    2019 , publisher=

    High-dimensional statistics: A non-asymptotic viewpoint , author=. 2019 , publisher=

    2019

  66. [74]

    Hanson-Wright inequality and sub-Gaussian concentration , author=

  67. [75]

    2018 , publisher=

    High-dimensional probability: An introduction with applications in data science , author=. 2018 , publisher=

    2018

  68. [76]

    Cryptology ePrint Archive , year=

    A note on sub-gaussian random variables , author=. Cryptology ePrint Archive , year=

  69. [77]

    Proceedings of the IEEE , volume=

    A comprehensive survey on transfer learning , author=. Proceedings of the IEEE , volume=. 2020 , publisher=

    2020

  70. [78]

    Journal of Big Data , volume=

    A survey on heterogeneous transfer learning , author=. Journal of Big Data , volume=. 2017 , publisher=

    2017

  71. [79]

    Introduction to the non-asymptotic analysis of random matrices , booktitle=

    Vershynin, Roman , editor=. Introduction to the non-asymptotic analysis of random matrices , booktitle=. 2012 , pages=

    2012

  72. [80]

    The Annals of Applied Statistics , volume=

    Targeting underrepresented populations in precision medicine: A federated transfer learning approach , author=. The Annals of Applied Statistics , volume=. 2023 , publisher=

    2023

  73. [81]

    arXiv preprint arXiv:2102.08503 , year=

    Federated evaluation and tuning for on-device personalization: System design & applications , author=. arXiv preprint arXiv:2102.08503 , year=

    arXiv

  74. [82]

    Apple , howpublished =

  75. [83]

    Nature Machine Intelligence , volume=

    Federated machine learning in data-protection-compliant research , author=. Nature Machine Intelligence , volume=. 2023 , publisher=

    2023

  76. [84]

    Nature Reviews Genetics , volume=

    Towards precision medicine , author=. Nature Reviews Genetics , volume=. 2016 , publisher=

    2016

  77. [85]

    New England journal of medicine , volume=

    A new initiative on precision medicine , author=. New England journal of medicine , volume=. 2015 , publisher=

    2015

  78. [86]

    arXiv preprint arXiv:2312.12786 , year=

    Heterogeneous Transfer Learning for Building High-Dimensional Generalized Linear Models with Disparate Datasets , author=. arXiv preprint arXiv:2312.12786 , year=

    arXiv

  79. [87]

    Information and Inference: A Journal of the IMA , volume=

    On the optimality of averaging in distributed statistical learning , author=. Information and Inference: A Journal of the IMA , volume=. 2016 , publisher=

    2016

  80. [88]

    Journal of the American Statistical Association , year=

    Communication-efficient distributed statistical inference , author=. Journal of the American Statistical Association , year=

  81. [89]

    Statistical Theory and Related Fields , volume=

    A review of distributed statistical inference , author=. Statistical Theory and Related Fields , volume=. 2022 , publisher=

    2022

  82. [90]

    IEEE communications surveys & tutorials , volume=

    Federated learning in mobile edge networks: A comprehensive survey , author=. IEEE communications surveys & tutorials , volume=. 2020 , publisher=

    2020

  83. [91]

    The Annals of Mathematical Statistics , pages=

    Consistency of the maximum likelihood estimator in the presence of infinitely many incidental parameters , author=. The Annals of Mathematical Statistics , pages=. 1956 , publisher=

    1956

  84. [92]

    The Annals of Statistics , volume=

    Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities , author=. The Annals of Statistics , volume=. 2001 , publisher=

    2001

  85. [93]

    arXiv preprint arXiv:1407.2635 , year=

    Nonparametric empirical Bayes and maximum likelihood estimation for high-dimensional data analysis , author=. arXiv preprint arXiv:1407.2635 , year=

    arXiv

  86. [94]

    Mathematical Programming , volume=

    A distributed one-step estimator , author=. Mathematical Programming , volume=. 2019 , publisher=

    2019

  87. [95]

    The Annals of Statistics , volume=

    On the nonparametric maximum likelihood estimator for Gaussian location mixture densities with application to Gaussian denoising , author=. The Annals of Statistics , volume=. 2020 , publisher=

    2020

  88. [96]

    The Journal of Machine Learning Research , volume=

    Communication-efficient algorithms for statistical optimization , author=. The Journal of Machine Learning Research , volume=. 2013 , publisher=

    2013

  89. [97]

    Journal of Econometrics , volume=

    The second-order bias and mean squared error of nonlinear estimators , author=. Journal of Econometrics , volume=. 1996 , publisher=

    1996

  90. [98]

    Advances in Neural Information Processing Systems , volume=

    Mean estimation with user-level privacy under data heterogeneity , author=. Advances in Neural Information Processing Systems , volume=

  91. [99]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Multivariate, heteroscedastic empirical bayes via nonparametric maximum likelihood , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 2025 , publisher=

    2025

  92. [100]

    2000 , publisher=

    Asymptotic statistics , author=. 2000 , publisher=

    2000

  93. [101]

    Bernoulli , volume=

    Berry--Esseen bounds for multivariate nonlinear statistics with applications to M-estimators and stochastic gradient descent algorithms , author=. Bernoulli , volume=. 2022 , publisher=

    2022

  94. [102]

    Probability Theory and Related Fields , volume=

    Stein’s method for normal approximation in Wasserstein distances with application to the multivariate central limit theorem , author=. Probability Theory and Related Fields , volume=. 2020 , publisher=

    2020

  95. [103]

    Lindsay , title =

    Surajit Ray and Bruce G. Lindsay , title =. The Annals of Statistics , number =

  96. [104]

    The Annals of Statistics , volume=

    Statistical inference for decentralized federated learning , author=. The Annals of Statistics , volume=. 2024 , publisher=

    2024

  97. [105]

    Journal of the American Statistical Association , number=

    Federated adaptive causal estimation (face) of target treatment effects , author=. Journal of the American Statistical Association , number=. 2025 , publisher=

    2025

  98. [106]

    Statistics in medicine , volume=

    Collaborative inference for treatment effect with distributed data-sharing management in multicenter studies , author=. Statistics in medicine , volume=. 2024 , publisher=

    2024

  99. [107]

    The Annals of Statistics , number =

    Wenhua Jiang and Cun-Hui Zhang , title =. The Annals of Statistics , number =

  100. [108]

    Advances in Neural Information Processing Systems , volume=

    Improving diffusion models for inverse problems using manifold constraints , author=. Advances in Neural Information Processing Systems , volume=

  101. [109]

    Advances in Neural Information Processing Systems , volume=

    Noise2score: tweedie’s approach to self-supervised image denoising without clean images , author=. Advances in Neural Information Processing Systems , volume=

  102. [110]

    arXiv preprint arXiv:2208.11970 , year=

    Understanding diffusion models: A unified perspective , author=. arXiv preprint arXiv:2208.11970 , year=

    arXiv

  103. [111]

    arXiv preprint arXiv:2310.06721 , year=

    Tweedie moment projected diffusions for inverse problems , author=. arXiv preprint arXiv:2310.06721 , year=

    arXiv

  104. [112]

    Journal of the American Statistical Association , volume=

    Bayesian inference using synthetic likelihood: asymptotics and adjustments , author=. Journal of the American Statistical Association , volume=. 2023 , publisher=

    2023

  105. [113]

    Bayesian Analysis , volume=

    Bayesian restricted likelihood methods: Conditioning on insufficient statistics in Bayesian regression (with discussion) , author=. Bayesian Analysis , volume=. 2021 , publisher=

    2021

  106. [114]

    Journal of the American Statistical Association , volume=

    Tweedie’s formula and selection bias , author=. Journal of the American Statistical Association , volume=. 2011 , publisher=

    2011

  107. [115]

    International Conference on Machine Learning , pages=

    Personalized federated learning under mixture of distributions , author=. International Conference on Machine Learning , pages=. 2023 , organization=

    2023

  108. [116]

    Advances in Neural Information Processing Systems , volume=

    Towards personalized federated learning via heterogeneous model reassembly , author=. Advances in Neural Information Processing Systems , volume=

  109. [117]

    Advances in Neural Information Processing Systems , volume=

    Personalized federated learning with mixture of models for adaptive prediction and model fine-tuning , author=. Advances in Neural Information Processing Systems , volume=

  110. [118]

    arXiv preprint arXiv:2212.14444 , year=

    Empirical Bayes when estimation precision predicts parameters , author=. arXiv preprint arXiv:2212.14444 , year=

  111. [119]

    Advances in Neural Information Processing Systems , volume=

    Fedpop: A bayesian approach for personalised federated learning , author=. Advances in Neural Information Processing Systems , volume=

  112. [120]

    Journal of the American Statistical Association , volume=

    Convex optimization, shape constraints, compound decisions, and empirical Bayes rules , author=. Journal of the American Statistical Association , volume=. 2014 , publisher=

    2014

  113. [121]

    Annals of Statistics , year=

    Convergence of latent mixing measures in finite and infinite mixture models , author=. Annals of Statistics , year=

  114. [122]

    2008 , publisher=

    Optimal transport: old and new , author=. 2008 , publisher=

    2008

  115. [123]

    1999 , publisher=

    Real analysis: modern techniques and their applications , author=. 1999 , publisher=

    1999

  116. [124]

    Interpolation between H

    Soudsk. Interpolation between H. Journal of Mathematical Analysis and Applications , volume=. 2018 , publisher=

    2018

  117. [125]

    2017 , publisher=

    Fundamentals of nonparametric Bayesian inference , author=. 2017 , publisher=

    2017

  118. [126]

    International conference on machine learning , pages=

    Scaffold: Stochastic controlled averaging for federated learning , author=. International conference on machine learning , pages=. 2020 , organization=

    2020

  119. [127]

    2014 IEEE 55th annual symposium on foundations of computer science , pages=

    Private empirical risk minimization: Efficient algorithms and tight error bounds , author=. 2014 IEEE 55th annual symposium on foundations of computer science , pages=. 2014 , organization=

    2014

  120. [128]

    2023 , eprint=

    Differentially private inference via noisy optimization , author=. 2023 , eprint=

    2023

  121. [129]

    The Annals of Statistics , volume=

    Differentially private inference via noisy optimization , author=. The Annals of Statistics , volume=. 2023 , publisher=

    2023

  122. [130]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Gaussian differential privacy , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 2022 , publisher=

    2022

  123. [131]

    The Stata Journal , volume=

    Instrumental variables and GMM: Estimation and testing , author=. The Stata Journal , volume=. 2003 , publisher=

    2003

  124. [132]

    Advances in neural information processing systems , volume=

    Inverting gradients-how easy is it to break privacy in federated learning? , author=. Advances in neural information processing systems , volume=

  125. [133]

    arXiv preprint arXiv:1710.06963 , year=

    Learning differentially private recurrent language models , author=. arXiv preprint arXiv:1710.06963 , year=

    arXiv

  126. [134]

    Proceedings of the 2016 ACM SIGSAC conference on computer and communications security , pages=

    Deep learning with differential privacy , author=. Proceedings of the 2016 ACM SIGSAC conference on computer and communications security , pages=

    2016

  127. [135]

    arXiv preprint arXiv:1712.07557 , year=

    Differentially private federated learning: A client level perspective , author=. arXiv preprint arXiv:1712.07557 , year=

    arXiv

  128. [136]

    International conference on artificial intelligence and statistics , pages=

    Differentially private federated learning on heterogeneous data , author=. International conference on artificial intelligence and statistics , pages=. 2022 , organization=

    2022

  129. [137]

    Computational Statistics & Data Analysis , volume=

    Communication-efficient distributed M-estimation with missing data , author=. Computational Statistics & Data Analysis , volume=. 2021 , publisher=

    2021

  130. [138]

    Scientific data , volume=

    The eICU Collaborative Research Database, a freely available multi-center database for critical care research , author=. Scientific data , volume=. 2018 , publisher=

    2018

  131. [139]

    Chest , volume=

    The search for optimal oxygen saturation targets in critically ill patients: observational data from large ICU databases , author=. Chest , volume=. 2020 , publisher=

    2020

  132. [140]

    Artificial intelligence and statistics , pages=

    Communication-efficient learning of deep networks from decentralized data , author=. Artificial intelligence and statistics , pages=. 2017 , organization=

    2017

  133. [141]

    Theory of cryptography conference , pages=

    Calibrating noise to sensitivity in private data analysis , author=. Theory of cryptography conference , pages=. 2006 , organization=

    2006

  134. [142]

    , title =

    LeCun, Yann and Cortes, Corinna and Burges, Christopher J.C. , title =. 1998 , howpublished =

    1998

  135. [143]

    An empirical

    Robbins, Herbert , journal=. An empirical. 1956 , publisher=

    1956

  136. [144]

    Journal of Machine Learning Research , volume=

    Distributed statistical inference under heterogeneity , author=. Journal of Machine Learning Research , volume=

  137. [145]

    arXiv preprint arXiv:2605.02070 , year=

    Sharp regret-Hellinger bounds for Gaussian empirical Bayes via polynomial approximation , author=. arXiv preprint arXiv:2605.02070 , year=

    arXiv

  138. [146]

    The van trees inequality in the spirit of h

    Gassiat, Elisabeth and Stoltz, Gilles , journal=. The van trees inequality in the spirit of h. 2024 , publisher=

    2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.