Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

Newfluence: Boosting Model interpretability and Understanding in High Dimensions

T0 review · 2 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read In high dimensions, the classical influence function undercounts true leave-one-out influence per data point, and Newfluence removes the bias at the same cost.

desk verdict Clean bias decomposition and a plausible fix for influence functions in high dimensions, but the main proof leans on an unverified lemma from the authors' own preprint. read the letter →

arxiv 2507.11895 v1 pith:O32QDTO7 submitted 2025-07-16 stat.ML cs.LGstat.ME

Haolin Zou , Arnab Auddy , Yongchan Kwon , Kamiar Rahnama Rad , Arian Maleki This is my paper
classification stat.MLcs.LGstat.ME MSC 62F3562J0762R07
keywords influencefunctionsleave-one-outhigh-dimensionalstatisticsgeneralizedlinearmodelsinterpretabilityNewtonstepWoodburyidentityregularizedempiricalriskminimization
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

Machine-learning interpretability tools routinely use influence functions to estimate how much removing one training point changes the prediction on a test point. The standard formula, which avoids retraining by making a first-order Taylor approximation around the fitted model, was designed for settings where the number of parameters is far smaller than the number of observations. This paper argues that in modern high-dimensional regimes, where the parameter count is comparable to the sample size, that formula develops a systematic per-data-point bias: it approximates $(1-H_{ii})$ times the true influence, so influential points can appear unimportant. The paper introduces Newfluence, a single-Newton-step estimator with the same computational cost that is shown to estimate the true leave-one-out influence consistently. If the argument is correct, a widely used attribution technique has a high-dimensional blind spot, and a simple correction closes it.

What carries the argument

The load-bearing object is a rank-one-corrected Newton step. Starting from the full-data fit $\hat\beta$, one Newton step toward the leave-one-out minimizer is $\tilde\beta^{\mathrm{Newton}}_{/i}=\hat\beta+\dot{\ell}_i(\hat\beta)G^{-1}x_i/(1-H_{ii})$, obtained by applying the Woodbury identity to the leave-one-out Hessian. Newfluence defines $I_{\mathrm{New}}(z_i,z_0)=\ell_0(\tilde\beta^{\mathrm{Newton}}_{/i})-\ell_0(\hat\beta)$. This carries the argument because it approximates the retrained prediction without retraining, and because the Woodbury derivation exposes exactly where the classical approximation goes wrong: the standard formula omits the $1/(1-H_{ii})$ factor and therefore reports $(1-H_{ii})I_{\mathrm{True}}$ instead of the true influence. The consistency proof additionally relies on a bound showing that the single Newton step is close to the true leave-one-out retrained estimate.

What would settle it

For square-loss linear regression with Gaussian features and fixed $n/p$, compute $I_{\mathrm{IF}}$, $H_{ii}$, and exact leave-one-out influence $I_{\mathrm{True}}$ for every training point. The paper's Part 1 and Part 3 imply $(1-H_{ii})^{-1}I_{\mathrm{IF}}-I_{\mathrm{True}}=o_P(\mathrm{PolyLog}(n)/n)$; if the residual stays of order $1/\sqrt{n}$ instead, the central claim is false. A simpler empirical falsifier is to rerun the paper's logistic-ridge simulation and check whether Newfluence's Kendall $\tau$ remains near 1 while classical IF falls to roughly 0.88.

Watch

Extended reading notes

Core claim

The paper's central claim is an asymptotic identity for generalized linear models with Gaussian features and a smooth strongly convex regularizer. As $n,p\to\infty$ with $n/p\to\gamma$, the classical influence estimate $I_{\mathrm{IF}}(z_i,z_0)$ equals $(1-H_{ii})\,I_{\mathrm{True}}(z_i,z_0)+o_P(\mathrm{PolyLog}(n)/n)$, where $H_{ii}=x_i^\top G^{-1}x_i\,\ddot{\ell}_i(\hat\beta)$ is the leverage-like diagonal entry of $XG^{-1}X^\top\mathrm{diag}(\ddot{\ell}_i)$ and $G$ is the Hessian of the regularized loss. Since $H_{ii}$ is typically of constant order, the classical estimator understates true influence, and the distortion varies from point to point. The paper's proposed Newfluence estimator evaluates the test loss at a single Newton step from the full-data solution, $\hat\beta + \dot{\ell}_i(\hat\beta)G^{-1}x_i/(1-H_{ii})$, and is proven to satisfy $|I_{\mathrm{New}}(z_i,z_0)-I_{\mathrm{True}}(z_i,z_0)|=o_P(\mathrm{PolyLog}(n)/n)$. In logistic ridge experiments at $n/p=1/2$ with weak regularization, Newfluence achieves near-perfect Kendall rank correlation with exact leave-one-out influence while the classical approximation drops to about 0.88.

Load-bearing premise

Everything depends on an unverified bound, taken from the authors' earlier preprint, that one Newton step lands within a tiny error of the true retrained model; if that bound is wrong, the main consistency result is unsupported.

Editorial extensions

If this is right

  • Influence-function rankings computed from the classical formula are not merely noisy in high dimensions; they are systematically biased, so data debugging, data valuation, and model-alignment decisions built on them should be rechecked.
  • Newfluence corrects the bias at essentially the same computational cost as the classical formula, since the same Hessian inverse is reused and each point needs only a rank-one update.
  • In low-dimensional or strongly regularized models, where $\mathrm{df}/p$ is small, classical influence functions remain accurate, and Newfluence matches them; the correction matters exactly when model complexity is non-negligible relative to the sample size.
  • The paper's high-dimensional asymptotic framework applies beyond influence functions and can be used to analyze other attribution tools such as Shapley values.
  • For logistic ridge models with $n/p=0.5$ and $\lambda=0.01$, the experiments report near-perfect Kendall $\tau\approx 1.00$ for Newfluence versus $\approx 0.88$ for classical influence functions.

Reading between the lines

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

  • The same $1/(1-H_{ii})$ correction should apply to other leave-one-out diagnostics built on influence-style approximations, such as fast estimates of prediction error or cross-validated risk, so the bias documented here may be more widespread than influence attribution alone.
  • Because $H_{ii}$ acts as a per-point leverage, the correction grows as the model approaches interpolation; the paper's strong-convexity assumption keeps $1-H_{ii}$ bounded away from zero, but behavior near the interpolation boundary is a natural place to stress-test the method.
  • The single-Newton-step construction is model-agnostic, so the correction could be adapted to non-convex models by applying the step in the last linear layer or a local quadratic surrogate, and then checked against brute-force leave-one-out on small networks.
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

2 major / 5 minor

Summary. The paper studies regularized empirical risk minimization in the proportional asymptotics n,p→∞ with n/p→γ. It claims that in this regime the classical influence function I_IF of Koh and Liang systematically underestimates the true leave-one-out influence I_True by a per-point factor 1−H_ii, with error o_P(PolyLog(n)/n), while the proposed Newfluence estimator, a one-step Newton correction using the Woodbury formula, satisfies |I_New−I_True|=o_P(PolyLog(n)/n) at the same computational cost. Theory is developed for GLMs with smooth, strongly convex separable regularizers and Gaussian features; simulations on logistic ridge regression show near-perfect Kendall τ for Newfluence and degraded τ for I_IF at λ=0.01, with both performing well at λ=10.

Significance. If Theorem 2.1 holds, this is a valuable and surprising result: the classical influence function carries a datapoint-dependent multiplicative bias that does not vanish when p is comparable to n, and the bias can be removed by a parameter-free correction requiring no retraining. The Woodbury-based derivation is clean, the estimator has no tuned parameters, and the experiments directly test the ranking claim. The principal weakness is that the proof of Part 1 rests on Lemma 3.3 of a same-author preprint that is not stated or proved here, so the significance is conditional on that external result being valid and its hypotheses matching A1-A4 and B1-B2.

major comments (2)
  1. [Appendix B.1 / Theorem 2.1] The proof of Part 1 is entirely based on Lemma 3.3 of Zou et al. (2025), a preprint by the same authors that is neither stated nor proved in this manuscript. That lemma supplies the key bound ||bβ/i − tildeβ^Newton/i|| = o_P(PolyLog(n)/√n), from which both the Newfluence consistency (Part 1) and the bias formula (Part 3) follow. If the lemma has extra hypotheses beyond A1-A4 and B1-B2—for example a lower bound on λ, a stronger Hessian-Lipschitz condition, or an initialization/basin assumption—the central claims are unsupported as stated. Please include the full lemma statement and proof, or an independent proof under A1-A4/B1-B2, before the theorem can be accepted.
  2. [Appendix A.1 / Remark 2.2] The claim 'we claim without proof that, for most commonly used regularizers we have ||∇^2 r(β)|| = O_P(1)' is used to show H_ii is bounded below and hence (1−H_ii)^{-1}=O_P(1) in the proof of Part 2. Remark 2.2 also asserts H_ii=Θ_P(1/(1+λ)) and max_j r¨_j(β_j)=O_P(1) without proof. These bounds are load-bearing for the nonvanishing-bias conclusion. Please prove them under the paper's assumptions or restrict the regularizer class (e.g., to ridge/elastic net with uniformly bounded Hessian); otherwise the factor 1−H_ii could be degenerate or the inverse unstable.
minor comments (5)
  1. [Section 2.1] The sentence 'by choosing γ < 1, we it will even cover the situation where the number of features are less than the number of observations' is garbled and the direction is wrong: γ<1 corresponds to p>n. Please fix.
  2. [Lemma C.4] In the proof of Lemma C.4, the definition of z should be z=√p x_i (or x_i/√p depending on convention), not z=n^{-1/2} x_i; as written the scaling does not produce the quoted χ² tail probability.
  3. [Appendix B.1] In the chain proving Part 1, the displayed inequality should use |x_0^T(·)| rather than x_0^T(·) before the Cauchy-Schwarz step, since the polynomial-growth factor multiplies an absolute value.
  4. [Notation / throughout] The symbol PolyLog(n) is used for different polynomials in different places. Please state whether the theorem's o_P(PolyLog(n)/n) bound holds for some fixed polynomial of log n or for every polynomial, and use consistent notation.
  5. [Section 4 / Abstract] The name is inconsistently typeset as NEWFLUENCE and Newfluence; please unify.

Circularity Check

1 steps flagged · score 4.0 of 10

One load-bearing self-citation: Theorem 2.1's consistency bound is imported from Lemma 3.3 of the same authors' preprint rather than proved here.

  1. self citation load bearing [Appendix B.1, Proof of Theorem 2.1 Part 1 (using Zou et al., 2025, Lemma 3.3)]
    "It follows from Lemma 3.3 of (Zou et al., 2025) (with m = t = 1) that ∥bβ/i − ˜βNewton/i∥ = oP (1√n PolyLog(n))."

    The central consistency claim of Theorem 2.1 (Part 1), which anchors Parts 2 and 3 and the paper's main message, is not established inside the manuscript. The key error bound for a single Newton step is taken from Lemma 3.3 of Zou et al. (2025), a preprint by the same five authors, and the lemma is neither stated nor proved in this paper. In the displayed chain leading to (11), the final decay rate is entirely 'the above quoted bound on the error of the Newton step.' Thus the theorem's oP(PolyLog(n)/n) conclusion reduces to the content of that self-citation; if the lemma's hypotheses are stricter than A1–A4/B1–B2 or its proof is wrong, the Newfluence consistency and the (1−Hii) IF-bias formula do not follow.

full rationale

The paper's Newfluence estimator is a parameter-free first-order Newton correction, and the derivation in Appendix A via the Woodbury formula is self-contained; there is no fitted parameter that is later called a prediction. The only significant circularity signal is the importation of Lemma 3.3 from Zou et al. (2025), an overlapping-author preprint, as the sole justification for the Newton-step error bound on which Theorem 2.1 Parts 1–3 depend. Because that lemma is not reproduced, machine-checked, or independently verified here, the central consistency proof is not self-contained at its load-bearing point. I score this 4 rather than 6 or 8: the central claim is not equivalent to its inputs by construction, and the cited lemma is a specific external (if same-author) statement that could in principle be checked; the rest of the Woodbury expansion, concentration bounds, and Hii analysis are internally coherent. The unproved heuristic bound in Appendix A.1 is a limitation but not circular.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central result rests on the paper's stated assumptions, standard linear algebra and concentration results, plus one externally cited lemma from the authors' own 2025 preprint. No constants are fitted to data. Newfluence is a new estimator, not an invented entity.

assumptions (7)
  • domain assumption A1-A4: separable, twice-differentiable, proper convex, nu-strongly convex regularizer; loss and regularizer have polynomial growth bounds and Lipschitz Hessian.
    Invoked in the statement of Theorem 2.1, Section 2.2. Excludes LASSO despite Section 2.1 mentioning it as an example.
  • domain assumption B1: features are i.i.d. Gaussian with covariance Sigma satisfying lambda_max(Sigma) <= CX/p.
    Used for all concentration bounds on ||x_i||, ||X||, and x_0, Section 2.2.
  • domain assumption B2: responses have polynomially bounded tails with PolyLog decay.
    Used in proof of Part 1, Appendix B.1, to bound loss derivatives.
  • ad hoc to paper Lemma 3.3 of Zou et al. (2025): Newton-step error ||beta_hat_/i - beta_tilde_/i^Newton|| = oP(1/sqrt(n) PolyLog(n)).
    External self-cited lemma is the key input to Part 1 of Theorem 2.1 and is not proven in this paper.
  • standard math Woodbury matrix identity (Lemma C.1).
    Used to invert the leave-one-out Hessian in Appendix A.
  • standard math Gaussian concentration bounds (Lemmas C.2-C.4).
    Used to control norms of the design matrix and features.
  • ad hoc to paper 0 <= Hii <= 1 and Hii = ThetaP(1/(1+lambda)) for common losses.
    Stated in Section 2.1 and Remark 2.2 without a complete proof; Hii <= 1 follows from projection algebra, but the ThetaP rate is asserted.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Newfluence: Boosting Model interpretability and Understanding in High Dimensions." pith.science (2026). https://pith.science/paper/O32QDTO7

@misc{pith2026250711895,
  author       = {Pith},
  title        = {Pith review of: Newfluence: Boosting Model interpretability and Understanding in High Dimensions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O32QDTO7}},
  note         = {Machine review of arXiv:2507.11895}
}
abstract

The increasing complexity of machine learning (ML) and artificial intelligence (AI) models has created a pressing need for tools that help scientists, engineers, and policymakers interpret and refine model decisions and predictions. Influence functions, originating from robust statistics, have emerged as a popular approach for this purpose. However, the heuristic foundations of influence functions rely on low-dimensional assumptions where the number of parameters $p$ is much smaller than the number of observations $n$. In contrast, modern AI models often operate in high-dimensional regimes with large $p$, challenging these assumptions. In this paper, we examine the accuracy of influence functions in high-dimensional settings. Our theoretical and empirical analyses reveal that influence functions cannot reliably fulfill their intended purpose. We then introduce an alternative approximation, called Newfluence, that maintains similar computational efficiency while offering significantly improved accuracy. Newfluence is expected to provide more accurate insights than many existing methods for interpreting complex AI models and diagnosing their issues. Moreover, the high-dimensional framework we develop in this paper can also be applied to analyze other popular techniques, such as Shapley values.

Figures

Figures reproduced from arXiv: 2507.11895 by the authors.

Figure 1
Figure 1. Comparison of Newfluence and I IF(zi, z0). We use logistic ridge regression with n = 500, p = 1000, and λ = 0.01. The other details of the simulation are presented in Section 3 The figure shows results for all the influence of all the n = 500 training points on the prediction loss of m = 100 unseen new test points generated from the true logistic model. Left: Newfluence vs. true influence. Right: I IF(zi, z0) vs. tr… view at source ↗

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. On the Accuracy of Newton Step and Influence Function Data Attributions

    cs.LG 2025-12 unverdicted novelty 7.0 of 10

    New analysis without global strong convexity yields tight scaling laws: NS error ~Θ(kd/n²) and NS-IF difference ~Θ((k+d)√(kd)/n²) for well-behaved logistic regressions.

Reference graph

Works this paper leans on

36 extracted references · 26 canonical work pages · cited by 1 Pith paper

  1. [1]

    write newline

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

  2. [2]

    Approximate leave-one-out cross validation for regression with l1 regularizers

    Auddy, A., Zou, H., Rahnama Rad, K., and Maleki, A. Approximate leave-one-out cross validation for regression with l1 regularizers. IEEE Transactions on Information Theory, 70 0 (11): 0 8040--8071, 2024

    work page 2024

  3. [3]

    Bae, J., Ng, N., Lo, A., Ghassemi, M., and Grosse, R. B. If influence functions are the answer, then what is the question? Advances in Neural Information Processing Systems, 35: 0 17953--17967, 2022

    work page 2022

  4. [4]

    Influence functions in deep learning are fragile

    Basu, S., Pope, P., and Feizi, S. Influence functions in deep learning are fragile. arXiv preprint arXiv:2006.14651, 2020

    arXiv 2006

  5. [5]

    and Montanari, A

    Donoho, D. and Montanari, A. High dimensional robust m-estimation: Asymptotic variance via approximate message passing. Probability Theory and Related Fields, 166: 0 935--969, 2016

    work page 2016

  6. [6]

    L., Maleki, A., and Montanari, A

    Donoho, D. L., Maleki, A., and Montanari, A. The noise-sensitivity phase transition in compressed sensing. IEEE Transactions on Information Theory, 57 0 (10): 0 6920--6941, 2011

    work page 2011

  7. [7]

    J., Lim, C., and Yu, B

    El Karoui, N., Bean, D., Bickel, P. J., Lim, C., and Yu, B. On robust regression with high-dimensional predictors. Proceedings of the National Academy of Sciences, 110 0 (36): 0 14557--14562, 2013

    work page 2013

  8. [8]

    and Zou, J

    Ghorbani, A. and Zou, J. Data shapley: Equitable valuation of data for machine learning. In International conference on machine learning, pp.\ 2242--2251. PMLR, 2019

    work page 2019

Show all 36 references
  1. [9]

    Studying large language model generalization with influence functions

    Grosse, R., Bae, J., Anil, C., Elhage, N., Tamkin, A., Tajdini, A., Steiner, B., Li, D., Durmus, E., Perez, E., et al. Studying large language model generalization with influence functions. arXiv preprint arXiv:2308.03296, 2023

    2023 arXiv

  2. [10]

    Simfluence: Modeling the influence of individual training examples by simulating training runs

    Guu, K., Webson, A., Pavlick, E., Dixon, L., Tenney, I., and Bolukbasi, T. Simfluence: Modeling the influence of individual training examples by simulating training runs. arXiv preprint arXiv:2303.08114, 2023

    2023 arXiv

  3. [11]

    and Lowd, D

    Hammoudeh, Z. and Lowd, D. Training data influence analysis and estimation: A survey. Machine Learning, 113 0 (5): 0 2351--2403, 2024

    2024

  4. [12]

    Hampel, F. R. The influence curve and its role in robust estimation. Journal of the american statistical association, 69 0 (346): 0 383--393, 1974

    1974

  5. [13]

    C., and Tsvetkov, Y

    Han, X., Wallace, B. C., and Tsvetkov, Y. Explaining black box predictions and unveiling data artifacts through influence functions. arXiv preprint arXiv:2005.06676, 2020

    2005 arXiv

  6. [14]

    M., Engstrom, L., Leclerc, G., and Madry, A

    Ilyas, A., Park, S. M., Engstrom, L., Leclerc, G., and Madry, A. Datamodels: Predicting predictions from training data. In Proceedings of the 39th International Conference on Machine Learning, 2022

    2022

  7. [15]

    and Maleki, A

    Jalali, S. and Maleki, A. New approach to bayesian high-dimensional linear regression. Information and Inference: A Journal of the IMA, 7, 07 2016

    2016

  8. [16]

    A., Hynes, N., G \"u rel, N

    Jia, R., Dao, D., Wang, B., Hubis, F. A., Hynes, N., G \"u rel, N. M., Li, B., Zhang, C., Song, D., and Spanos, C. J. Towards efficient data valuation based on the shapley value. In The 22nd International Conference on Artificial Intelligence and Statistics, pp.\ 1167--1176. P...

    2019

  9. [17]

    Koh, P. W. and Liang, P. Understanding black-box predictions via influence functions. In Precup, D. and Teh, Y. W. (eds.), Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pp.\ 1885--1894. PMLR, 06--11...

    2017

  10. [18]

    and Zou, J

    Kwon, Y. and Zou, J. Beta shapley: a unified and noise-reduced data valuation framework for machine learning. In International Conference on AI and Statistics, 2022

    2022

  11. [19]

    Datainf: Efficiently estimating data influence in lora-tuned llms and diffusion models, 2024

    Kwon, Y., Wu, E., Wu, K., and Zou, J. Datainf: Efficiently estimating data influence in lora-tuned llms and diffusion models, 2024. URL https://arxiv.org/abs/2310.00902

    2024 arXiv

  12. [20]

    and Wei, Y

    Li, Y. and Wei, Y. Minimum _1 -norm interpolators: Precise asymptotics and multiple descent. arXiv preprint arXiv:2110.09502, 2021

    2021 arXiv

  13. [21]

    Understanding impact of human feedback via influence functions

    Min, T., Lee, H., Ryu, H., Kwon, Y., and Lee, K. Understanding impact of human feedback via influence functions. arXiv preprint arXiv:2501.05790, 2025

    2025 arXiv

  14. [22]

    M., Georgiev, K., Ilyas, A., Leclerc, G., and Madry, A

    Park, S. M., Georgiev, K., Ilyas, A., Leclerc, G., and Madry, A. Trak: Attributing model behavior at scale. In International Conference on Machine Learning, pp.\ 27074--27113. PMLR, 2023

    2023

  15. [23]

    Estimating training data influence by tracing gradient descent

    Pruthi, G., Liu, F., Kale, S., and Sundararajan, M. Estimating training data influence by tracing gradient descent. Advances in Neural Information Processing Systems, 33: 0 19920--19930, 2020

    2020

  16. [24]

    and Maleki, A

    Rahnama Rad, K. and Maleki, A. A scalable estimate of the out-of-sample prediction error via approximate leave-one-out cross-validation. Journal of the Royal Statistical Society Series B: Statistical Methodology, 82 0 (4): 0 965--996, 2020

    2020

  17. [25]

    Error bounds in estimating the out-of-sample prediction error using leave-one-out cross validation in high-dimensions

    Rahnama Rad, K., Zhou, W., and Maleki, A. Error bounds in estimating the out-of-sample prediction error using leave-one-out cross validation in high-dimensions. In International Conference on Artificial Intelligence and Statistics, pp.\ 4067--4077. PMLR, 2020

    2020

  18. [26]

    The shapley value in machine learning

    Rozemberczki, B., Watson, L., Bayer, P., Yang, H.-T., Kiss, O., Nilsson, S., and Sarkar, R. The shapley value in machine learning. In The 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence, pp.\ 5572--5579...

    2022

  19. [27]

    Theoretical and practical perspectives on what influence functions do

    Schioppa, A., Filippova, K., Titov, I., and Zablotskaia, P. Theoretical and practical perspectives on what influence functions do. Advances in Neural Information Processing Systems, 36: 0 27560--27581, 2023

    2023

  20. [28]

    and Najmi, A

    Sundararajan, M. and Najmi, A. The many shapley values for model explanation. In International conference on machine learning, pp.\ 9269--9278. PMLR, 2020

    2020

  21. [29]

    Sur, P., Chen, Y., and Cand \`e s, E. J. The likelihood ratio test in high-dimensional logistic regression is asymptotically a rescaled chi-square. Probability theory and related fields, 175: 0 487--558, 2019

    2019

  22. [30]

    T., Mittal, P., Song, D., and Jia, R

    Wang, J. T., Mittal, P., Song, D., and Jia, R. Data shapley in one training run. arXiv preprint arXiv:2406.11011, 2024 a

    2024 arXiv

  23. [31]

    T., Song, D., Zou, J., Mittal, P., and Jia, R

    Wang, J. T., Song, D., Zou, J., Mittal, P., and Jia, R. Capturing the temporal dependence of training data influence. arXiv preprint arXiv:2412.09538, 2024 b

    2024 arXiv

  24. [32]

    Approximate leave-one-out for fast parameter tuning in high dimensions

    Wang, S., Zhou, W., Lu, H., Maleki, A., and Mirrokni, V. Approximate leave-one-out for fast parameter tuning in high dimensions. In International Conference on Machine Learning, pp.\ 5228--5237. PMLR, 2018

    2018

  25. [33]

    I., and Ravikumar, P

    Yeh, C.-K., Hsieh, C.-Y., Suggala, A., Inouye, D. I., and Ravikumar, P. K. On the (in) fidelity and sensitivity of explanations. Advances in neural information processing systems, 32, 2019

    2019

  26. [34]

    Correcting large language model behavior via influence function

    Zhang, H., Zhang, Z., Zhang, Y., Zhai, Y., Peng, H., Lei, Y., Yu, Y., Wang, H., Liang, B., Gui, L., et al. Correcting large language model behavior via influence function. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pp.\ 14477--14485, 2025

    2025

  27. [35]

    Does _p -minimization outperform _1 -minimization? IEEE Transactions on Information Theory, 63 0 (11): 0 6896--6935, 2017

    Zheng, L., Maleki, A., Weng, H., Wang, X., and Long, T. Does _p -minimization outperform _1 -minimization? IEEE Transactions on Information Theory, 63 0 (11): 0 6896--6935, 2017

    2017

  28. [36]

    Certified data removal under high-dimensional settings

    Zou, H., Auddy, A., Kwon, Y., Rahnama Rad, K., and Maleki, A. Certified data removal under high-dimensional settings. arXiv preprint arXiv:2505.07640, 2025. URL https://arxiv.org/abs/2505.07640

    2025 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.