pith. sign in

arxiv: 2605.04317 · v2 · pith:CGWCRL4Enew · submitted 2026-05-05 · 🧮 math.ST · stat.TH

The Threshold Breakdown Point

Tianjun Ke , Marco Avella Medina This is my paper

Pith reviewed 2026-05-20 23:48 UTC · model grok-4.3

classification 🧮 math.ST stat.TH
keywords robust statisticsbreakdown pointM-estimatorsfinite sample analysishypothesis testingasymptotic normalitybootstrap inference
0
0 comments X

The pith

The threshold breakdown point measures the smallest contamination fraction needed to force a prescribed deviation in an estimator.

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

This paper defines the threshold breakdown point as the minimal fraction of contaminated observations required to push an estimator past a chosen deviation level. It pairs this with the finite sample m-sensitivity, which gives the largest deviation an estimator can suffer when exactly m points are replaced. The authors derive both quantities for standard M-estimators along with their standard errors and test statistics. The same framework extends earlier decision breakdown ideas to hypothesis testing and supplies asymptotic normality plus a multiplier bootstrap for inference on these new measures.

Core claim

The paper establishes that the threshold breakdown point and m-sensitivity admit explicit derivations for M-estimators under ordinary regularity conditions, furnishing finite-sample robustness diagnostics that correspond to the breakdown functions previously studied only in the asymptotic regime and enabling bootstrap-based uncertainty statements for both estimation and testing.

What carries the argument

The threshold breakdown point, the smallest contamination fraction that produces a user-specified deviation from the uncontaminated estimator value.

If this is right

  • The measures extend directly to standard errors and test statistics, yielding breakdown characterizations for hypothesis tests.
  • They serve as finite-sample analogues of the power and level breakdown functions studied in earlier asymptotic work.
  • Consistency and asymptotic normality hold for the threshold breakdown and m-sensitivity under the same regularity conditions used for the original estimators.
  • A multiplier bootstrap supplies valid uncertainty quantification without additional assumptions.

Where Pith is reading between the lines

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

  • The same construction could be applied to other estimator families once their worst-case contamination maps are derived.
  • Practitioners might use the threshold to set minimum sample sizes that keep expected deviation below a tolerance.
  • Links between the finite-sample m-sensitivity and classical influence functions remain open for further study.

Load-bearing premise

The explicit formulas rest on a contamination model that allows direct calculation of the worst-case deviation together with standard differentiability and uniqueness conditions for the M-estimator objective.

What would settle it

Compute the threshold breakdown point for a concrete M-estimator on a fixed sample; if replacing that exact fraction of points never produces a deviation as large as the prescribed value across repeated trials, the claimed threshold is incorrect.

Figures

Figures reproduced from arXiv: 2605.04317 by Marco Avella Medina, Tianjun Ke.

Figure 1
Figure 1. Figure 1: The threshold breakdown points and m-sensitivity for the two-stage M-estimator under N (0, 1), Cauchy(0, 1), and Unif(0, 1) distributions with n = 1000, m ∈ {10, 20, . . . , 150}, and η ∈ {0.1, 0.2, . . . , 1}, using Huber’s loss for location and Huber’s proposal 2 for scale (Huber, 1964). The figure suggests a tail-dependent finite sample effect: when m is sufficiently large for contamination to interact … view at source ↗
Figure 2
Figure 2. Figure 2: Upper and lower bounds for the rejection breakdown point BPreject(ϕ, x(n) ) as a function of the effect size θ under N (θ, 1). Line types distinguish n ∈ {500, 1000, 2000}; the shaded band indicates the interval between the lower and upper bounds. the bound gap is smallest around weak signals, where the inequalities in Theorem 11 are empirically tightest for these procedures. The contrast between the score… view at source ↗
Figure 3
Figure 3. Figure 3: Sensitivity curves for the calcium–placebo blood pressure data. Panels (a)–(c) report the m-sensitivity ηm/n(θb x(nx) −θb y (ny) ) for Huber’s, logcosh, and self-concordant losses, respectively. The horizontal axis is the discrete level m/ min(nx, ny); at m/ min(nx, ny) = 0.5 complete breakdown occurs, matching the classical finite sample regime. Red markers (and the dashed line connecting them) are the co… view at source ↗
Figure 4
Figure 4. Figure 4: Standardized test statistic bands for the calcium-placebo blood pressure data. For each loss, we report with the light-blue bars, at level m/ min(nx, ny), the lower bound and upper bound given by Theorem 11 for (θbx˜ − θby˜)/se( b θbx˜, θby˜), where ˜x and ˜y denote the contaminated datasets at each level. Gray dashed lines are again one-sided thresholds z1−α. A bar lying entirely above a gray line certifi… view at source ↗
Figure 5
Figure 5. Figure 5: Loss functions (left) and the corresponding score functions (right) for Huber’s, logcosh, and self-concordant losses. The tuning parameters are set to δ = 1.345 for Huber, δ = 1.2047 for logcosh, and δ = 1.4811 for the self-concordant loss. 0 100 200 300 400 m 0 2 4 6 8 Normal Cauchy Uniform (a) Huber’s loss. 0 100 200 300 400 m 0 2 4 6 8 Normal Cauchy Uniform (b) Logcosh loss. 0 100 200 300 400 m 0 2 4 6 … view at source ↗
Figure 6
Figure 6. Figure 6: m-sensitivity ηm/n(θ, x b (n) ) for location M-estimators. Top panels: n = 1000 with m ∈ {20, 60, . . . , 460}, and line types distinguish data generating distributions. Bottom panels: m/n ∈ {0.02, 0.06, . . . , 0.46}, and line types distinguish sample sizes under N (0, 1). 42 view at source ↗
Figure 7
Figure 7. Figure 7: Gap between the upper and lower bounds on BPreject(ϕ, x(n) ) as a function of n, with θ = 1. Points are Monte Carlo replications; line and (almost invisible) band are a fitted regression and its 95% confidence band. Line types distinguish the generating distribution. 2.0 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 Effect Size 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 Breakdown Point of Rejection 50 100 200 (a) Hube… view at source ↗
Figure 8
Figure 8. Figure 8: Upper and lower bounds for the rejection breakdown point BPreject(ϕ, x(nx) , y(ny) ) as a function of the effect size θ in the two-sample problem. Line types distinguish nx = ny ∈ {50, 100, 200}; the shaded band indicates the interval between the lower and upper bounds. 44 view at source ↗
Figure 9
Figure 9. Figure 9: Randomized PIT–based uniformity diagnostics for the multiplier bootstrap. Each panel corresponds to a different trimming proportion ε ∈ {0.03, 0.1, 0.15}, with n = 100, M = 1000, and B = 1000. Under a valid bootstrap approximation, we expect the distribution to be uniform. Panels show PP curves near y = x and histograms near the y = 1 baseline, with in-panel Kolmogorov– Smirnov statistics and p-values indi… view at source ↗
Figure 10
Figure 10. Figure 10: Randomized PIT–based uniformity diagnostics for the multiplier bootstrap. Each panel corresponds to a different trimming proportion ε ∈ {0.03, 0.1, 0.15}, with n = 1000, M = 3000, and B = 3000. The finite sample deviation decreases compared to previous plots. 45 view at source ↗
read the original abstract

We introduce a novel approach to finite sample robustness that avoids the pessimism of traditional breakdown analyses. We define the threshold breakdown point, the smallest contamination fraction needed to induce a prescribed deviation, and the finite sample m-sensitivity, the worst-case deviation that an estimator can incur after m observations are contaminated. We derive these measures for commonly used M-estimators, their standard errors and related test statistics. This allows us to extend the decision breakdown point of Zhang (1996) to obtain general breakdown characterizations for hypothesis testing, and show how these notions correspond to finite sample counterparts of the power and level breakdown functions of He, Simpson and Portnoy (1990). We complement our work with an inferential framework for the threshold breakdown and m-sensitivity that yields consistency and asymptotic normality results, as well as a valid multiplier bootstrap for uncertainty quantification. We illustrate the practical utility of our methods in various numerical examples and an application to a two sample testing problem for a blood pressure dataset.

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

2 major / 2 minor

Summary. The paper introduces the threshold breakdown point (the smallest contamination fraction inducing a prescribed deviation from the uncontaminated estimator) and the finite-sample m-sensitivity (the worst-case deviation after m contaminated observations). These are derived for common M-estimators, their standard errors, and test statistics; the decision breakdown point of Zhang (1996) is extended to hypothesis testing and linked to the power and level breakdown functions of He, Simpson and Portnoy (1990). An inferential framework is provided that claims consistency, asymptotic normality, and a valid multiplier bootstrap for these new functionals, with numerical illustrations and an application to a two-sample blood-pressure test.

Significance. If the central asymptotic claims hold, the work supplies a less pessimistic, tunable finite-sample robustness measure that directly quantifies the contamination level needed to reach a user-specified deviation; the explicit link to existing breakdown functions for testing is a useful bridge to the literature. The multiplier bootstrap for uncertainty quantification on the threshold breakdown point itself would be a practical addition if the regularity conditions transfer.

major comments (2)
  1. [§4] §4 (Asymptotic theory for the threshold breakdown point): the argument that the min-over-contamination functional inherits the differentiability, strict convexity, and unique-minimizer properties required for standard M-estimator consistency and asymptotic normality is not supplied. Because the threshold breakdown point is itself defined via an inner optimization over contamination, it is unclear whether the effective objective remains differentiable at the threshold or satisfies the Lipschitz conditions used for the ordinary M-estimator; this is load-bearing for the claimed normality and bootstrap validity.
  2. [Theorem 5.1] Theorem 5.1 (multiplier bootstrap validity): the proof sketch invokes the same regularity conditions as the uncontaminated M-estimator, yet no verification is given that the worst-case deviation map preserves the moment and smoothness assumptions when the contamination fraction is the parameter being estimated. If the bootstrap weights interact with the inner contamination optimization, the exchangeability argument may fail.
minor comments (2)
  1. [Abstract] The abstract states that 'standard regularity conditions' are used but never enumerates them; adding a short list (e.g., differentiability of the objective, uniqueness of the minimizer, finite second moments) would improve readability.
  2. [Numerical examples] In the numerical examples section, the caption of Figure 3 does not specify the contamination model or the value of the prescribed deviation δ used to compute the threshold breakdown point.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive feedback on our manuscript. The comments on the asymptotic theory and bootstrap validity are well-taken, and we address them point by point below. We will revise the paper to supply the missing arguments and clarifications.

read point-by-point responses

  1. Referee: [§4] §4 (Asymptotic theory for the threshold breakdown point): the argument that the min-over-contamination functional inherits the differentiability, strict convexity, and unique-minimizer properties required for standard M-estimator consistency and asymptotic normality is not supplied. Because the threshold breakdown point is itself defined via an inner optimization over contamination, it is unclear whether the effective objective remains differentiable at the threshold or satisfies the Lipschitz conditions used for the ordinary M-estimator; this is load-bearing for the claimed normality and bootstrap validity.

    Authors: We agree that an explicit verification of property inheritance is required and was not sufficiently detailed in the original submission. In the revised manuscript we will insert a new supporting lemma in §4 establishing that, under the maintained strict convexity, continuous differentiability, and unique-minimizer assumptions on the underlying M-estimator loss, the threshold breakdown point functional (defined via the inner infimum over contamination) remains strictly convex and directionally differentiable at the threshold value. The Lipschitz condition is preserved by the continuous dependence of the worst-case deviation on the contamination fraction, which follows from the compactness of the contamination neighborhood and the uniform continuity of the loss. This lemma directly justifies the application of standard M-estimator consistency and asymptotic normality results to the threshold breakdown point. revision: yes

  2. Referee: [Theorem 5.1] Theorem 5.1 (multiplier bootstrap validity): the proof sketch invokes the same regularity conditions as the uncontaminated M-estimator, yet no verification is given that the worst-case deviation map preserves the moment and smoothness assumptions when the contamination fraction is the parameter being estimated. If the bootstrap weights interact with the inner contamination optimization, the exchangeability argument may fail.

    Authors: The referee correctly identifies that the preservation of regularity conditions under the worst-case deviation map must be verified explicitly. We will expand the proof of Theorem 5.1 with an additional proposition showing that the map from contamination fraction to worst-case deviation is Lipschitz continuous and preserves the required moment bounds (e.g., finite second moments of the score) under our standing assumptions on the loss function. Regarding the bootstrap, the multiplier weights are applied to the outer functional after the threshold has been estimated; the inner contamination optimization is treated as a fixed (data-dependent) map that does not interact with the weights. Consequently, the exchangeability of the bootstrap weights conditional on the observed sample continues to hold. We will add this clarification together with a short appendix lemma. revision: yes

Circularity Check

0 steps flagged

No circularity detected; definitions and asymptotics are constructed from explicit contamination model and standard M-estimator regularity conditions.

full rationale

The paper defines the threshold breakdown point directly as the smallest contamination fraction inducing a prescribed deviation and m-sensitivity as the worst-case deviation after m contaminated observations. These are derived for M-estimators under stated differentiability, uniqueness of minimizer, and moment conditions together with an explicit contamination model permitting closed-form worst-case calculations. Asymptotic consistency, normality, and multiplier bootstrap results are obtained by applying the same standard regularity conditions to the new functionals. Extensions reference external works (Zhang 1996; He, Simpson and Portnoy 1990) whose authors do not overlap with the present paper. No equation reduces a reported prediction or first-principles result to a fitted parameter or self-citation by construction; the central claims retain independent content from the definitions and external theory.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claims rest on standard regularity conditions for M-estimators and a contamination model that allows explicit worst-case calculations; no free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Standard regularity conditions for M-estimators (differentiability, uniqueness of minimizer, moment conditions)
    Invoked to derive consistency, asymptotic normality, and bootstrap validity of the new robustness measures.

pith-pipeline@v0.9.0 · 5686 in / 1384 out tokens · 39252 ms · 2026-05-20T23:48:38.282571+00:00 · methodology

Review history (2 revisions) →

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

121 extracted references · 121 canonical work pages · 3 internal anchors

  1. [1]

    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 , organization=

    work page 2016

  2. [2]

    Bias robustness of depth estimators in multivariate settings

    Some insights into depth estimators for location and scatter in the multivariate setting , author=. arXiv preprint arXiv:2505.07383 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  3. [3]

    Weighted

    Rosenbaum, Paul R , journal=. Weighted. 2014 , publisher=

    work page 2014

  4. [4]

    Annals of Mathematical Statistics , volume=

    A general qualitative definition of robustness , author=. Annals of Mathematical Statistics , volume=. 1971 , publisher=

    work page 1971

  5. [5]

    1986 , publisher=

    Robust Statistics: The Approach Based on Influence Functions , author=. 1986 , publisher=

    work page 1986

  6. [6]

    arXiv preprint arXiv:2002.08774 , year=

    Propose, test, release: Differentially private estimation with high probability , author=. arXiv preprint arXiv:2002.08774 , year=

    work page arXiv 2002

  7. [7]

    Symposium on Theory of Computing (STOC) , pages=

    Smooth sensitivity and sampling in private data analysis , author=. Symposium on Theory of Computing (STOC) , pages=

    work page

  8. [8]

    1968 , publisher=

    Contributions to the theory of robust estimation , author=. 1968 , publisher=

    work page 1968

  9. [9]

    A simple two-sample

    Wang, Min and Liu, Guangying , journal=. A simple two-sample. 2016 , publisher=

    work page 2016

  10. [10]

    Journal of the American Statistical Association , volume=

    Robust bounded-influence tests in general parametric models , author=. Journal of the American Statistical Association , volume=. 1994 , publisher=

    work page 1994

  11. [11]

    Econometrica , volume=

    Estimation and inference with weak, semi-strong, and strong identification , author=. Econometrica , volume=. 2012 , publisher=

    work page 2012

  12. [12]

    Biometrika , volume=

    Robust estimation of high-dimensional covariance and precision matrices , author=. Biometrika , volume=. 2018 , publisher=

    work page 2018

  13. [13]

    International Conference on Machine Learning (ICML) , volume=

    Convergence rates for differentially private statistical estimation , author=. International Conference on Machine Learning (ICML) , volume=

    work page

  14. [14]

    Bootstrap consistency for general semiparametric

    Cheng, Guang and Huang, Jianhua Z , journal=. Bootstrap consistency for general semiparametric. 2010 , publisher=

    work page 2010

  15. [15]

    Bernoulli (to appear) , year=

    On the breakdown point of transport-based quantiles , author=. Bernoulli (to appear) , year=

    work page

  16. [16]

    arXiv preprint, arXiv:2603.16005 , year=

    Breakdown properties of optimal transport maps: general transportation costs , author=. arXiv preprint, arXiv:2603.16005 , year=

    work page arXiv

  17. [17]

    A theoretical framework for

    Marusic, Juraj and Medina, Marco Avella and Rush, Cynthia , journal=. A theoretical framework for

    work page

  18. [18]

    Annals of Applied Probability (to appear) , year=

    On the robustness of semi-discrete optimal transport , author=. Annals of Applied Probability (to appear) , year=

    work page

  19. [19]

    Existence and breakdown analysis of

    Konen, Dimitri and Paindaveine, Davy , journal=. Existence and breakdown analysis of. 2025 , publisher=

    work page 2025

  20. [20]

    Annales de l’Institut Henri Poincar

    On the robustness of spatial quantiles , author=. Annales de l’Institut Henri Poincar

    work page

  21. [21]

    Asymptotic behaviour of

    Davies, Laurie P , journal=. Asymptotic behaviour of. 1987 , publisher=

    work page 1987

  22. [22]

    Journal of the American Statistical Association , volume=

    Least median of squares regression , author=. Journal of the American Statistical Association , volume=. 1984 , publisher=

    work page 1984

  23. [23]

    Robust regression by means of

    Rousseeuw, Peter and Yohai, Victor , booktitle=. Robust regression by means of. 1984 , organization=

    work page 1984

  24. [24]

    Annals of Statistics , pages=

    High breakdown-point and high efficiency robust estimates for regression , author=. Annals of Statistics , pages=. 1987 , publisher=

    work page 1987

  25. [25]

    Annals of Statistics , volume=

    On robustness and local differential privacy , author=. Annals of Statistics , volume=. 2023 , publisher=

    work page 2023

  26. [26]

    Privately estimating a

    Alabi, Daniel and Kothari, Pravesh K and Tankala, Pranay and Venkat, Prayaag and Zhang, Fred , booktitle=. Privately estimating a

    work page

  27. [27]

    Annals of Statistics , volume=

    Maximum bias curves for robust regression with non-elliptical regressors , author=. Annals of Statistics , volume=. 2001 , publisher=

    work page 2001

  28. [28]

    Journal of the american statistical association , volume=

    Wald's test as applied to hypotheses in logit analysis , author=. Journal of the american statistical association , volume=. 1977 , publisher=

    work page 1977

  29. [29]

    Symposium on Theory of Computing (STOC) , pages=

    Robustness implies privacy in statistical estimation , author=. Symposium on Theory of Computing (STOC) , pages=

    work page

  30. [30]

    Chance , volume=

    The role of robust statistics in private data analysis , author=. Chance , volume=. 2020 , publisher=

    work page 2020

  31. [31]

    Journal of the American Statistical Association , volume=

    Privacy-preserving parametric inference: a case for robust statistics , author=. Journal of the American Statistical Association , volume=. 2021 , publisher=

    work page 2021

  32. [32]

    Annals of Statistics , volume=

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

    work page 2023

  33. [33]

    Journal of Computational and Graphical Statistics , pages=

    Differentially private significance tests for regression coefficients , author=. Journal of Computational and Graphical Statistics , pages=. 2019 , publisher=

    work page 2019

  34. [34]

    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=

    work page 2014

  35. [35]

    Journal of the American Statistical Association , volume=

    Prepivoting test statistics: a bootstrap view of asymptotic refinements , author=. Journal of the American Statistical Association , volume=. 1988 , publisher=

    work page 1988

  36. [36]

    Transactions of the american mathematical society , volume=

    The accuracy of the Gaussian approximation to the sum of independent variates , author=. Transactions of the american mathematical society , volume=. 1941 , publisher=

    work page 1941

  37. [37]

    Advances in Neural Information Processing Systems (NeurIPS) , volume=

    Covariance-aware private mean estimation without private covariance estimation , author=. Advances in Neural Information Processing Systems (NeurIPS) , volume=

    work page

  38. [38]

    IEEE Transactions on Information Theory , volume=

    Bandits with heavy tail , author=. IEEE Transactions on Information Theory , volume=. 2013 , publisher=

    work page 2013

  39. [39]

    Advances in Neural Information Processing Systems (NeruIPS) , volume=

    Private hypothesis selection , author=. Advances in Neural Information Processing Systems (NeruIPS) , volume=

    work page

  40. [40]

    The cost of privacy: Optimal rates of convergence for parameter estimation with differential privacy,

    The cost of privacy: optimal rates of convergence for paramer estimaion with differential privacy , author=. arXiv preprint arXiv:1902.04495 , year=

    work page arXiv 1902

  41. [41]

    Annales de l'IHP Probabilit

    Challenging the empirical mean and empirical variance: a deviation study , author=. Annales de l'IHP Probabilit

    work page

  42. [42]

    Journal of Machine Learning Research , volume=

    Differentially private empirical risk minimization , author=. Journal of Machine Learning Research , volume=

    work page

  43. [43]

    International Conference on Machine Learning (ICML) , year=

    Convergence rates for differentially private statistical estimation , author=. International Conference on Machine Learning (ICML) , year=

    work page

  44. [44]

    The Annals of Statistics , volume=

    Robust covariance and scatter matrix estimation under Huber’s contamination model , author=. The Annals of Statistics , volume=. 2018 , publisher=

    work page 2018

  45. [45]

    Journal of Nonparametric Statistics , volume=

    Maxbias curves of robust location estimators based on subranges , author=. Journal of Nonparametric Statistics , volume=. 2002 , publisher=

    work page 2002

  46. [46]

    Journal of Econometrics , volume=

    The power of bootstrap and asymptotic tests , author=. Journal of Econometrics , volume=. 2006 , publisher=

    work page 2006

  47. [47]

    The Annals of Statistics , volume=

    Breakdown and groups , author=. The Annals of Statistics , volume=

    work page

  48. [48]

    REVSTAT-Statistical Journal , volume=

    The breakdown point—examples and counterexamples , author=. REVSTAT-Statistical Journal , volume=

    work page

  49. [49]

    An automatic finite-sample robustness metric: when can dropping a little data make a big difference?arXiv preprint arXiv:2011.14999,

    An automatic finite-sample robustness metric: when can dropping a little data make a big difference? , author=. arXiv preprint arXiv:2011.14999 , year=

    work page arXiv 2011

  50. [50]

    International Conference on Artificial Intelligence and Statistics (AISTATS) , pages=

    Influence diagnostics under self-concordance , author=. International Conference on Artificial Intelligence and Statistics (AISTATS) , pages=. 2023 , organization=

    work page 2023

  51. [51]

    Advances in neural information processing systems (NeurIPS) , volume=

    On the accuracy of influence functions for measuring group effects , author=. Advances in neural information processing systems (NeurIPS) , volume=

    work page

  52. [52]

    Advances in Neural Information Processing Systems (NeurIPS) , volume=

    Most influential subset selection: Challenges, promises, and beyond , author=. Advances in Neural Information Processing Systems (NeurIPS) , volume=

    work page

  53. [53]

    Statistical Science , volume=

    Robustness by reweighting for kernel estimators: an overview , author=. Statistical Science , volume=. 2021 , publisher=

    work page 2021

  54. [54]

    The Annals of Statistics , volume=

    Sub-Gaussian mean estimators , author=. The Annals of Statistics , volume=. 2016 , publisher=

    work page 2016

  55. [55]

    SIAM Journal on Computing , volume=

    Robust estimators in high-dimensions without the computational intractability , author=. SIAM Journal on Computing , volume=. 2019 , publisher=

    work page 2019

  56. [56]

    1997 , publisher=

    Evaluating density forecasts , author=. 1997 , publisher=

    work page 1997

  57. [57]

    A Festschrift for Erich L

    The notion of breakdown point , author=. A Festschrift for Erich L. Lehmann , year=

    work page

  58. [58]

    Journal of the American Statistical Association , volume=

    Minimax optimal procedures for locally private estimation , author=. Journal of the American Statistical Association , volume=. 2018 , publisher=

    work page 2018

  59. [59]

    , author=

    Differential privacy and robust statistics. , author=. Symposium on Theory of Computing (STOC) , volume=

    work page

  60. [60]

    Theory of cryptography conference , pages=

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

    work page 2006

  61. [61]

    Foundations and Trends

    The algorithmic foundations of differential privacy , author=. Foundations and Trends. 2014 , publisher=

    work page 2014

  62. [62]

    ICML'16 Proceedings of the 33rd International Conference on International Conference on Machine Learning-Volume 48 , year=

    Differentially private chi-squared hypothesis testing: Goodness of fit and independence testing , author=. ICML'16 Proceedings of the 33rd International Conference on International Conference on Machine Learning-Volume 48 , year=

    work page

  63. [63]

    Advances in Neural Information Processing Systems (NeurIPS) , volume=

    Privacy induces robustness: Information-computation gaps and sparse mean estimation , author=. Advances in Neural Information Processing Systems (NeurIPS) , volume=

    work page

  64. [64]

    , title =

    Hampel, Frank R. , title =. Journal of the American Statistical Association , year =

    work page

  65. [65]

    Journal of the American Statistical Association , volume=

    Breakdown robustness of tests , author=. Journal of the American Statistical Association , volume=. 1990 , publisher=

    work page 1990

  66. [66]

    Symposium on Theory of Computing (STOC) , pages=

    Efficient mean estimation with pure differential privacy via a sum-of-squares exponential mechanism , author=. Symposium on Theory of Computing (STOC) , pages=

    work page

  67. [67]

    The Journal of Machine Learning Research , volume=

    Loss minimization and parameter estimation with heavy tails , author=. The Journal of Machine Learning Research , volume=. 2016 , publisher=

    work page 2016

  68. [68]

    1994 , publisher=

    Robust Asymptotic Statistics , author=. 1994 , publisher=

    work page 1994

  69. [69]

    1982 , publisher=

    Robust testing in linear models: the infinitesimal approach , author=. 1982 , publisher=

    work page 1982

  70. [70]

    and Ronchetti, Elvezio , Edition =

    Huber, Peter J. and Ronchetti, Elvezio , Edition =. Robust Statistics , Year =

    work page

  71. [71]

    Annals of Mathematical Statistics , volume=

    Robust Estimation of a Location Parameter , author=. Annals of Mathematical Statistics , volume=. 1964 , publisher=

    work page 1964

  72. [72]

    The Annals of Statistics , volume=

    Finite Sample Breakdown of M -and P -Estimators , author=. The Annals of Statistics , volume=. 1984 , publisher=

    work page 1984

  73. [73]

    Tukey's contributions to robust statistics , author=

    John W. Tukey's contributions to robust statistics , author=. Annals of statistics , pages=. 2002 , publisher=

    work page 2002

  74. [74]

    Biometrika , volume=

    A simple resampling method by perturbing the minimand , author=. Biometrika , volume=. 2001 , publisher=

    work page 2001

  75. [75]

    Conference on Learning Theory (COLT) , pages=

    Private mean estimation of heavy-tailed distributions , author=. Conference on Learning Theory (COLT) , pages=

    work page

  76. [76]

    The Annals of Statistics , volume=

    Inference using noisy degrees: Differentially private beta-model and synthetic graphs , author=. The Annals of Statistics , volume=. 2016 , publisher=

    work page 2016

  77. [77]

    Statistical Science , volume=

    User-friendly covariance estimation for heavy-tailed distributions , author=. Statistical Science , volume=. 2019 , publisher=

    work page 2019

  78. [78]

    Working paper , year=

    The threshold breakdown point , author=. Working paper , year=

    work page

  79. [79]

    Conference on Learning Theory , pages=

    Private convex empirical risk minimization and high-dimensional regression , author=. Conference on Learning Theory , pages=

    work page

  80. [80]

    2008 , publisher=

    Introduction to empirical processes and semiparametric inference , author=. 2008 , publisher=

    work page 2008

Showing first 80 references.