pith. sign in

arxiv: 2606.00717 · v1 · pith:7GBSBYMAnew · submitted 2026-05-30 · 💻 cs.LG · cs.AI· stat.ML

Multi-Agent Conformal Prediction with Personalized Statistical Validity

Martin V. Vejling , Christophe A. N. Biscio , Adrien Mazoyer , Petar Popovski , Shashi Raj Pandey This is my paper

Pith reviewed 2026-06-28 19:11 UTC · model grok-4.3

classification 💻 cs.LG cs.AIstat.ML
keywords conformal predictionfederated learningmulti-agent systemsuncertainty quantificationdensity ratio estimationprivacy preservationasymptotic coverage
0
0 comments X

The pith

Personalized federated conformal prediction yields asymptotically valid coverage for each agent despite data heterogeneity.

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

The paper develops a multi-agent conformal prediction approach to handle privacy constraints, limited local data, and non-identical distributions across agents. It introduces personalized federated weighted conformal prediction that applies local density ratio weighting and then performs weighted quantile aggregation. The framework is designed to deliver valid marginal and calibration-conditional coverage guarantees individually per agent rather than only in aggregate. A reader would care because many real distributed systems require reliable uncertainty estimates without centralizing sensitive calibration data. The analysis also supplies an effective sample size expression that adjusts coverage variance under weighting.

Core claim

PFWCP combines local density ratio weighting with weighted quantile aggregation to correct for heterogeneity while preserving privacy. The method yields asymptotically valid marginal and calibration-conditional coverage guarantees for each participating agent and supports protocols with one-shot communication. Theoretical analysis presents an adjustment to the coverage variance governed by an effective sample size expression necessary in weighted conformal prediction.

What carries the argument

The PFWCP framework that integrates local density ratio weighting to correct heterogeneity with weighted quantile aggregation to maintain validity and privacy.

If this is right

  • Each agent obtains valid prediction sets without sharing raw calibration data.
  • Both marginal and calibration-conditional coverage hold asymptotically for every agent individually.
  • Only one round of communication is required for the protocol.
  • Coverage variance is characterized by an effective sample size that accounts for the weighting scheme.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method suggests density ratio weighting as a general tool for restoring validity in other federated uncertainty quantification settings.
  • Finite-sample behavior under varying degrees of heterogeneity remains an open question left implicit by the asymptotic analysis.
  • The one-shot communication property may enable deployment in bandwidth-constrained or latency-sensitive multi-agent systems.

Load-bearing premise

Local density ratio estimates accurately correct for data heterogeneity and the weighted quantile aggregation preserves validity in the asymptotic regime under the paper's weighting scheme.

What would settle it

An experiment in which the empirical coverage for one or more individual agents falls materially below the nominal level under strong heterogeneity would falsify the asymptotic per-agent guarantees.

Figures

Figures reproduced from arXiv: 2606.00717 by Adrien Mazoyer, Christophe A. N. Biscio, Martin V. Vejling, Petar Popovski, Shashi Raj Pandey.

Figure 1
Figure 1. Figure 1: Outline of the proposed methodology with comparison to the closest existing works. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Diagrams illustrating the communication signaling required to execute the PFWCP and [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Empirical CDF of the CCC for CP with covariate shift (blue solid line), CP without [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Boxplots of coverage and efficiency for the proposed methods and benchmarks on the [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
read the original abstract

Uncertainty quantification is essential in high-stakes machine learning tasks. However, one of the principled solutions, conformal prediction, faces challenges under limited local calibration data, privacy constraints, and data heterogeneity. In multi-agent settings, existing works do not simultaneously and satisfactorily address these challenges with guarantees either limited to averages across agents or losing validity in heterogeneous settings. Hence, we propose personalized federated weighted conformal prediction (PFWCP), a framework that combines local density ratio weighting with weighted quantile aggregation to correct for heterogeneity while preserving privacy. The method yields asymptotically valid marginal and calibration-conditional coverage guarantees for each participating agent and supports protocols with one-shot communication. Theoretical analysis presents an adjustment to the coverage variance, governed by an effective sample size expression, which is necessary in the context of weighted conformal prediction, and experiments on synthetic and real datasets show improved calibration quality over state-of-the-art federated conformal baselines.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper proposes Personalized Federated Weighted Conformal Prediction (PFWCP), a multi-agent framework that uses local density ratio weighting combined with weighted quantile aggregation to achieve asymptotically valid marginal and calibration-conditional coverage guarantees for each individual agent. It supports one-shot communication protocols while addressing data heterogeneity and privacy constraints, and includes a theoretical adjustment to coverage variance via an effective sample size expression, with experiments showing improved calibration over federated conformal baselines.

Significance. If the asymptotic per-agent guarantees hold under the stated conditions, the work would be significant for enabling personalized uncertainty quantification in federated settings with limited local data and heterogeneity, where prior methods either provide only average guarantees or lose validity. The effective-sample-size variance adjustment for weighted conformal prediction is a potentially useful technical contribution if the supporting rates are established.

major comments (1)
  1. [Theoretical analysis] Theoretical analysis (and abstract claim of asymptotic validity): the central per-agent calibration-conditional coverage guarantee requires that local density ratio estimators converge at a rate sufficient for the effective-sample-size adjustment to control the coverage error under the one-shot protocol. The manuscript does not identify or verify an explicit rate condition that the restricted one-shot information satisfies; without this, the conditional guarantee may fail asymptotically even if marginal coverage holds.
minor comments (2)
  1. Notation for the weighting scheme and effective sample size expression should be introduced with explicit definitions before the variance adjustment is stated, to improve readability of the theoretical claims.
  2. Experiments section: clarify whether the reported improvements are statistically significant across multiple random seeds and heterogeneous data partitions, and include a direct comparison of coverage error rates against the theoretical bound.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback. We respond to the single major comment below.

read point-by-point responses

  1. Referee: [Theoretical analysis] Theoretical analysis (and abstract claim of asymptotic validity): the central per-agent calibration-conditional coverage guarantee requires that local density ratio estimators converge at a rate sufficient for the effective-sample-size adjustment to control the coverage error under the one-shot protocol. The manuscript does not identify or verify an explicit rate condition that the restricted one-shot information satisfies; without this, the conditional guarantee may fail asymptotically even if marginal coverage holds.

    Authors: We agree that the current theoretical section assumes consistency of the local density ratio estimators without stating an explicit rate condition sufficient to ensure the calibration-conditional coverage error vanishes under the one-shot protocol. The marginal coverage result holds under milder conditions, but the conditional guarantee relies on the effective-sample-size term controlling the remainder, which does require a rate on the estimation error. In the revision we will add an explicit assumption (sup-norm error of each density ratio estimator = o_p(n_eff^{-1/2})) together with a short argument showing that the one-shot protocol can meet this rate when the shared model parameters permit consistent estimation at the required speed. This addition will be placed immediately before the statement of the conditional coverage theorem. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained against external benchmarks

full rationale

The abstract and description present asymptotic marginal and calibration-conditional coverage guarantees derived from local density ratio weighting combined with weighted quantile aggregation, plus an explicit effective-sample-size variance adjustment described as a necessary theoretical correction for weighted conformal prediction. No equations or steps are shown reducing the coverage claims to fitted inputs by construction, self-definitional loops, or load-bearing self-citations whose cited results themselves depend on the target result. The one-shot protocol and per-agent guarantees rest on stated assumptions about estimator convergence rather than tautological renaming or ansatz smuggling. This is the normal case of an independent theoretical derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim depends on the ability to estimate density ratios that correctly adjust for heterogeneity and on the validity of the asymptotic analysis for weighted conformal prediction; no explicit free parameters or invented entities are named in the abstract.

axioms (2)
  • domain assumption Density ratios between local and other agents' distributions can be estimated sufficiently accurately to enable valid reweighting
    Invoked to correct for heterogeneity while preserving privacy.
  • domain assumption The asymptotic regime applies such that coverage guarantees hold as sample sizes increase
    Basis for the marginal and calibration-conditional validity claims.

pith-pipeline@v0.9.1-grok · 5699 in / 1238 out tokens · 27093 ms · 2026-06-28T19:11:09.456433+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

109 extracted references · 9 canonical work pages · 1 internal anchor

  1. [1]

    2025 , eprint=

    Full-conformal novelty detection: A powerful and non-random approach , author=. 2025 , eprint=

    2025

  2. [2]

    2024 , eprint=

    Boosting e-BH via conditional calibration , author=. 2024 , eprint=

    2024

  3. [3]

    Proceedings of the 37th International Conference on Neural Information Processing Systems , articleno =

    Bashari, Meshi and Epstein, Amir and Romano, Yaniv and Sesia, Matteo , title =. Proceedings of the 37th International Conference on Neural Information Processing Systems , articleno =. 2023 , publisher =

    2023

  4. [4]

    Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications , pages=

    Enhancing Conformal Prediction Using E-Test Statistics , author=. Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications , pages=. 2024 , volume=

    2024

  5. [5]

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

    Ruodu Wang and Aaditya Ramdas , title =. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume =. 2022 , month =

    2022

  6. [6]

    Transactions on Machine Learning Research , issn=

    E-Valuating Classifier Two-Sample Tests , author=. Transactions on Machine Learning Research , issn=

  7. [7]

    2025 , eprint=

    E-Values Expand the Scope of Conformal Prediction , author=. 2025 , eprint=

    2025

  8. [8]

    2024 , eprint=

    Continuous Testing: Unifying Tests and E-values , author=. 2024 , eprint=

    2024

  9. [9]

    2025 , eprint=

    Real-time Program Evaluation using Anytime-valid Rank Tests , author=. 2025 , eprint=

    2025

  10. [10]

    2025 , eprint=

    Optimal Conformal Prediction, E-values, Fuzzy Prediction Sets and Subsequent Decisions , author=. 2025 , eprint=

    2025

  11. [11]

    Hypothesis Testing with E-values.Foundations and Trends in Statistics, Vol

    Foundations and Trends in Statistics , title =. 2025 , volume =. doi:10.1561/3600000002 , issn =

    work page doi:10.1561/3600000002 2025

  12. [12]

    Proceedings of The 27th International Conference on Artificial Intelligence and Statistics , pages =

    Online multiple testing with e-values , author =. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics , pages =. 2024 , editor =

    2024

  13. [13]

    Yifan Zhang and Zijian Wei and Haojie Ren and Changliang Zou , booktitle=. e-. 2025 , url=

    2025

  14. [14]

    The Annals of Statistics , number =

    Stephen Bates and Emmanuel Cand. The Annals of Statistics , number =. 2023 , doi =

    2023

  15. [15]

    The Annals of Statistics , number =

    Ariane Marandon and Lihua Lei and David Mary and Etienne Roquain , title =. The Annals of Statistics , number =. 2024 , doi =

    2024

  16. [16]

    Selection by prediction with conformal p-values , year =

    Jin, Ying and Cand\`. Selection by prediction with conformal p-values , year =. Journal of Machine Learning Research , month =

  17. [17]

    The Thirteenth International Conference on Learning Representations , year=

    Conditional Testing based on Localized Conformal \ p\ -values , author=. The Thirteenth International Conference on Learning Representations , year=

  18. [18]

    2025 , eprint=

    Unified Conformalized Multiple Testing with Full Data Efficiency , author=. 2025 , eprint=

    2025

  19. [19]

    2025 , eprint=

    Sequentializing a Test: Anytime Validity is Free , author=. 2025 , eprint=

    2025

  20. [20]

    2025 , eprint=

    An online generalization of the (e-)Benjamini-Hochberg procedure , author=. 2025 , eprint=

    2025

  21. [21]

    Proceedings of the 20th International Conference on Artificial Intelligence and Statistics , pages =

    Communication-Efficient Learning of Deep Networks from Decentralized Data , author =. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics , pages =. 2017 , volume =

    2017

  22. [22]

    The Annals of Statistics , number =

    Jia Gu and Song Xi Chen , title =. The Annals of Statistics , number =. 2024 , doi =

    2024

  23. [23]

    and Raskar, Ramesh , title =

    Lu, Charles and Yu, Yaodong and Karimireddy, Sai Praneeth and Jordan, Michael I. and Raskar, Ramesh , title =. proceedings of the 40th International Conference on Machine Learning , articleno =. 2023 , publisher =

    2023

  24. [24]

    Federated Inference With Reliable Uncertainty Quantification Over Wireless Channels via Conformal Prediction , year=

    Zhu, Meiyi and Zecchin, Matteo and Park, Sangwoo and Guo, Caili and Feng, Chunyan and Simeone, Osvaldo , journal=. Federated Inference With Reliable Uncertainty Quantification Over Wireless Channels via Conformal Prediction , year=

  25. [25]

    proceedings of the 40th International Conference on Machine Learning , pages =

    One-Shot Federated Conformal Prediction , author =. proceedings of the 40th International Conference on Machine Learning , pages =. 2023 , volume =

    2023

  26. [26]

    proceedings of the 27th International Conference on Artificial Intelligence and Statistics , pages =

    Efficient Conformal Prediction under Data Heterogeneity , author =. proceedings of the 27th International Conference on Artificial Intelligence and Statistics , pages =. 2024 , volume =

    2024

  27. [27]

    Proceedings of the 41st International Conference on Machine Learning , pages =

    Multi-Source Conformal Inference Under Distribution Shift , author =. Proceedings of the 41st International Conference on Machine Learning , pages =. 2024 , volume =

    2024

  28. [28]

    Statistical Theory and Related Fields , volume =

    Yuan Gao and Weidong Liu and Hansheng Wang and Xiaozhou Wang and Yibo Yan and Riquan Zhang , title =. Statistical Theory and Related Fields , volume =. 2022 , publisher =

    2022

  29. [29]

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

    Sesia, Matteo and Wang, Y X Rachel and Tong, Xin , title =. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume =. 2024 , month =

    2024

  30. [30]

    proceedings of the 40th International Conference on Machine Learning , pages =

    Conformal Prediction for Federated Uncertainty Quantification Under Label Shift , author =. proceedings of the 40th International Conference on Machine Learning , pages =. 2023 , volume =

    2023

  31. [31]

    The Annals of Statistics , number =

    Adel Javanmard and Andrea Montanari , title =. The Annals of Statistics , number =. 2018 , doi =

    2018

  32. [32]

    Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications , pages =

    Conformal Predictive Systems Under Covariate Shift , author =. Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications , pages =. 2024 , editor =

    2024

  33. [33]

    Conformal Prediction Under Covariate Shift , volume =

    Tibshirani, Ryan J and Foygel Barber, Rina and Candes, Emmanuel and Ramdas, Aaditya , booktitle =. Conformal Prediction Under Covariate Shift , volume =

  34. [34]

    proceedings of the 39th Annual Conference on Neural Information Processing Systems , year=

    Personalized Federated Conformal Prediction with Localization , author=. proceedings of the 39th Annual Conference on Neural Information Processing Systems , year=

  35. [35]

    2025 , eprint=

    Probably Approximately Correct Labels , author=. 2025 , eprint=

    2025

  36. [36]

    The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

    Robust Conformal Prediction Using Privileged Information , author=. The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

  37. [37]

    2025 , eprint=

    Conformal Prediction with Corrupted Labels: Uncertain Imputation and Robust Re-weighting , author=. 2025 , eprint=

    2025

  38. [38]

    Conference on Uncertainty in Artificial Intelligence , pages =

    Distribution-free uncertainty quantification for classification under label shift , author =. Conference on Uncertainty in Artificial Intelligence , pages =. 2021 , volume =

    2021

  39. [39]

    Proceedings of The 28th International Conference on Artificial Intelligence and Statistics , pages =

    Conformal Prediction Under Generalized Covariate Shift with Posterior Drift , author =. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics , pages =. 2025 , volume =

    2025

  40. [40]

    Proceedings of the 42nd International Conference on Machine Learning , pages =

    Robust Conformal Outlier Detection under Contaminated Reference Data , author =. Proceedings of the 42nd International Conference on Machine Learning , pages =. 2025 , volume =

    2025

  41. [41]

    Machine Learning , number =

    Cherubin, Giovanni , title =. Machine Learning , number =. 2019 , doi =

    2019

  42. [42]

    2022 , issn =

    Multi split conformal prediction , journal =. 2022 , issn =. doi:10.1016/j.spl.2022.109395 , author =

    work page doi:10.1016/j.spl.2022.109395 2022

  43. [43]

    Journal of the American Statistical Association , volume =

    Yachong Yang and Arun Kumar Kuchibhotla , title =. Journal of the American Statistical Association , volume =. 2025 , publisher =

    2025

  44. [44]

    Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli, and Rob J Hyndman

    Nested conformal prediction and quantile out-of-bag ensemble methods , journal =. 2022 , issn =. doi:10.1016/j.patcog.2021.108496 , author =

    work page doi:10.1016/j.patcog.2021.108496 2022

  45. [45]

    2024 , eprint=

    Marginal and training-conditional guarantees in one-shot federated conformal prediction , author =. 2024 , eprint=

    2024

  46. [46]

    Forty-second International Conference on Machine Learning , year=

    False Coverage Proportion Control for Conformal Prediction , author=. Forty-second International Conference on Machine Learning , year=

  47. [47]

    Second Workshop on Test-Time Adaptation: Putting Updates to the Test! at ICML 2025 , year=

    An Evidence-Based Post-Hoc Adjustment Framework for Anomaly Detection Under Data Contamination , author=. Second Workshop on Test-Time Adaptation: Putting Updates to the Test! at ICML 2025 , year=

    2025

  48. [48]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    One Sample is Enough to Make Conformal Prediction Robust , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  49. [49]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    Conformal Prediction under L\'evy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  50. [50]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    Backward Conformal Prediction , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  51. [51]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    Class conditional conformal prediction for multiple inputs by p-value aggregation , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  52. [52]

    The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

    Adaptive Latent-Space Constraints in Personalized Federated Learning , author=. The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=

  53. [53]

    Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications , pages =

    On the Calibration of Aggregated Conformal Predictors , author =. Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications , pages =. 2017 , editor =

    2017

  54. [54]

    Artificial Intelligence Applications and Innovations , year=

    Carlsson, Lars and Eklund, Martin and Norinder, Ulf , title=. Artificial Intelligence Applications and Innovations , year=

  55. [55]

    Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications , pages =

    Combination of Conformal Predictors for Classification , author =. Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications , pages =. 2017 , volume =

    2017

  56. [56]

    2024 , eprint=

    Merging uncertainty sets via majority vote , author=. 2024 , eprint=

    2024

  57. [57]

    2025 , eprint=

    Theoretical Foundations of Conformal Prediction , author=. 2025 , eprint=

    2025

  58. [58]

    Veeravalli and Anirban Roy and Susmit Jha , title =

    Akshayaa Magesh and Venugopal V. Veeravalli and Anirban Roy and Susmit Jha , title =. Journal of Machine Learning Research , year =

  59. [59]

    Advances in Neural Information Processing Systems , editor =

    Adaptive Conformal Inference Under Distribution Shift , author =. Advances in Neural Information Processing Systems , editor =

  60. [60]

    Isaac Gibbs and Emmanuel J. Cand. Conformal Inference for Online Prediction with Arbitrary Distribution Shifts , journal =. 2024 , volume =

    2024

  61. [61]

    Oliveira and Paulo Orenstein and Thiago Ramos and Jo

    Roberto I. Oliveira and Paulo Orenstein and Thiago Ramos and Jo. Split Conformal Prediction and Non-Exchangeable Data , journal =. 2024 , volume =

    2024

  62. [62]

    The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

    Personalized Federated Learning via Feature Distribution Adaptation , author=. The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

  63. [63]

    Proceedings of the 38th International Conference on Machine Learning , pages =

    Ditto: Fair and Robust Federated Learning Through Personalization , author =. Proceedings of the 38th International Conference on Machine Learning , pages =. 2021 , editor =

    2021

  64. [64]

    NeurIPS 2024 Workshop on Bayesian Decision-making and Uncertainty , year=

    Two Students: Enabling Uncertainty Quantification in Federated Learning Clients , author=. NeurIPS 2024 Workshop on Bayesian Decision-making and Uncertainty , year=

    2024

  65. [65]

    Hong-You Chen and Wei-Lun Chao , booktitle=. Fed. 2021 , url=

    2021

  66. [66]

    Fed-ensemble: Ensemble Models in Federated Learning for Improved Generalization and Uncertainty Quantification , year=

    Shi, Naichen and Lai, Fan and Al Kontar, Raed and Chowdhury, Mosharaf , journal=. Fed-ensemble: Ensemble Models in Federated Learning for Improved Generalization and Uncertainty Quantification , year=

  67. [67]

    The Thirteenth International Conference on Learning Representations , year=

    Unlocking the Potential of Model Calibration in Federated Learning , author=. The Thirteenth International Conference on Learning Representations , year=

  68. [68]

    2025 , issn =

    Federated learning meets Bayesian neural network: Robust and uncertainty-aware distributed variational inference , journal =. 2025 , issn =. doi:10.1016/j.neunet.2025.107135 , author =

    work page doi:10.1016/j.neunet.2025.107135 2025

  69. [69]

    2026 , eprint=

    Tighter confidence intervals for quantiles of heterogeneous data , author=. 2026 , eprint=

    2026

  70. [70]

    2024 , eprint=

    Federated Prediction-Powered Inference from Decentralized Data , author=. 2024 , eprint=

    2024

  71. [71]

    2025 , eprint=

    Extending Prediction-Powered Inference through Conformal Prediction , author=. 2025 , eprint=

    2025

  72. [72]

    FedCF: Fair Federated Conformal Prediction

    Anutam Srinivasan and Aditya T. Vadlamani and Amin Meghrazi and Srinivasan Parthasarathy , year=. 2509.22907 , archivePrefix=

    work page internal anchor Pith review Pith/arXiv arXiv

  73. [73]

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

    Gibbs, Isaac and Cherian, John J and Candès, Emmanuel J , title =. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume =. 2025 , month =

    2025

  74. [74]

    2020 , editor =

    Karimireddy, Sai Praneeth and Kale, Satyen and Mohri, Mehryar and Reddi, Sashank and Stich, Sebastian and Suresh, Ananda Theertha , booktitle =. 2020 , editor =

    2020

  75. [75]

    Angelopoulos and Stephen Bates and Clara Fannjiang and Michael I

    Anastasios N. Angelopoulos and Stephen Bates and Clara Fannjiang and Michael I. Jordan and Tijana Zrnic , title =. Science , volume =. 2023 , doi =

    2023

  76. [76]

    Proceedings of the Asian Conference on Machine Learning , pages =

    Conditional Validity of Inductive Conformal Predictors , author =. Proceedings of the Asian Conference on Machine Learning , pages =. 2012 , editor =

    2012

  77. [77]

    Semi-Supervised Risk Control via Prediction-Powered Inference , year=

    Einbinder, Bat-Sheva and Ringel, Liran and Romano, Yaniv , journal=. Semi-Supervised Risk Control via Prediction-Powered Inference , year=

  78. [78]

    The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

    Active, anytime-valid risk controlling prediction sets , author=. The Thirty-eighth Annual Conference on Neural Information Processing Systems , year=

  79. [79]

    Proceedings of The 26th International Conference on Artificial Intelligence and Statistics , pages =

    Improving Adaptive Conformal Prediction Using Self-Supervised Learning , author =. Proceedings of The 26th International Conference on Artificial Intelligence and Statistics , pages =. 2023 , editor =

    2023

  80. [80]

    2025 , eprint=

    Semi-Supervised Conformal Prediction With Unlabeled Nonconformity Score , author=. 2025 , eprint=

    2025

Showing first 80 references.