Pith. sign in

REVIEW 2 major objections 7 minor 300 references

Gradient-free stochastic optimization of derivatives under strong convexity

T0 review · 2 major / 7 minor · reviewed 2026-07-09 · glm-5.2

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

Pith's one-line read Optimal rate found for minimizing derivatives from noisy queries

desk verdict First minimax analysis for optimizing higher-order derivatives from sequential zero-order queries; matching bounds in N, open dimension gap. read the letter →

arxiv 2607.07249 v1 pith:DTSFHTTI submitted 2026-07-08 math.ST stat.TH

Arya Akhavan , Sirine Louati , Alexandre B. Tsybakov This is my paper · ORCID
classification math.STstat.TH MSC 62L2090C5662G05
keywords zero-orderoptimizationstochasticminimaxlowerboundskernelestimationHöldersmoothnessderivativestrongconvexityprojectedgradient
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper solves the problem of finding the minimizer of the k-th partial derivative f = ∂_j^k g of an unknown function g, when the learner can only query g at sequentially chosen points and receives noisy values. The main result is that the minimax optimal rate of the optimization error E[f(x̂_N) − f*] is exactly N^{−(β−1)/(β+k)} in the total number of queries N, where β ≥ 2 is the Hölder smoothness of f and k ≥ 0 is the order of differentiation. The upper bound is achieved by a kernel-based projected stochastic gradient algorithm using 2d paired queries per iteration with a carefully tuned smoothing parameter h ∝ T^{−1/(2(β+k))}, and the matching lower bound holds over all sequential algorithms, establishing that each additional order of differentiation costs exactly one unit of smoothness in the convergence rate exponent.

What carries the argument

The proof of the upper bound combines: (i) a kernel-based gradient estimator (equation 9) using two bounded kernels K_k and K_1 with specific moment-cancellation properties (7)–(8), paired perturbations along e_j and e_l, and h^{−(k+1)} scaling; (ii) a bias bound of order h^{β−1} (Lemma 5.1) exploiting Hölder regularity and kernel moment cancellation; (iii) a second-moment bound of order G^2 + d h^{2(β−1)} + d h^{−2(k+1)} (Lemma 5.2) capturing noise, smoothness of g, and the h^{−(k+1)} amplification; (iv) a Chung-type recursion (Lemma 6.1) for the sequence Δ_t = E[||x_t − x*||^2] with two forcing terms A/(t+t_0) and B/(t+t_0)^2, where the A-term encodes squared bias and the B-term encodesvar

What would settle it

If one could exhibit a sequential algorithm achieving optimization error o(N^{−(β−1)/(β+k)}) uniformly over the class F'_{α,β,k}, Theorem 5.2 would be falsified. Conversely, if the lower-bound construction's two hypotheses g_± failed to satisfy all seven properties (a)–(g) of the class F'_{α,β,k} simultaneously for the chosen parameters, the lower bound would not apply.

Watch

Extended reading notes

Core claim

The central discovery is the exact minimax rate N^{−(β−1)/(β+k)} for zero-order optimization of a k-th order derivative under Hölder smoothness β+k and strong convexity of f. The rate is achieved by a two-kernel gradient estimator: one kernel K_k extracts the k-th derivative along the perturbation coordinate j (using moment-cancellation conditions (7)), and a second kernel K_1 extracts the first-order derivative along each gradient coordinate l (using moment-cancellation conditions (8)). The paired perturbations hUe_j ± hVe_l and the scaling h^{−(k+1)} recover ∂_l ∂_j^k g(x_t) = ∂_l f(x_t) up to a bias of order h^{β−1} and a variance of order h^{−2(k+1)}. The optimal bias-variance trade-off,

Load-bearing premise

The entire argument relies on f = ∂_j^k g being α-strongly convex on Θ, which ensures the minimizer is unique and well-separated, drives the linear convergence factor α/2 in the recursion, and allows the lower bound to convert squared distance into optimization error. Without strong convexity, the minimizer need not be unique and neither the upper-bound recursion nor the lower-bound construction would go through.

Editorial extensions

If this is right

  • The rate N^{−(β−1)/(β+k)} quantifies precisely how much harder it is to optimize a derivative than the function itself: each order of differentiation k shifts the exponent down by 1/(β+k) relative to the direct case, establishing a smoothness cost calculus for indirect zero-order optimization.
  • The two-kernel construction (K_k for the derivative direction, K_1 for the gradient direction) provides a template for estimating gradients of composite functionals from zero-order data, potentially extending to mixed partial derivatives and other linear functionals of g.
  • The gap between the upper bound's dimension factor d^{(2β+k−1)/(β+k)} and the dimension-free lower bound identifies the sharp dimension dependence as the main open problem, analogous to the direct case (k=0) where the exponent 2−1/β is known to be reducible to 1 for β=2,3.
  • The sequential query structure allows rates depending on d only polynomially, in contrast to passive (i.i.d.) observation where rates depend exponentially on d—a separation between active and passive regimes that extends to derivative optimization.

Reading between the lines

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

  • The smoothness cost of k derivatives being exactly k units in the rate exponent suggests a general principle: for zero-order optimization of any linear functional L[g] that costs s units of smoothness to estimate, the optimal rate should be N^{−(β−1−s)/(β+s)} under analogous assumptions, though this extrapolation is not established in the paper.
  • The lower-bound construction via two hypotheses g_± = q ± r h^{β+k} Φ(x_1/h) that are indistinguishable under KL divergence ≤ 1/8 yet have k-th derivatives with well-separated minimizers could potentially extend to mixed partial derivatives, though the authors note the two-kernel construction is tailored to a single coordinate direction.
  • Relaxing strong convexity to mere convexity would likely change the achievable rate, since the α/2 factor in the recursion (30) and the lower-bound reduction (67) both depend on strong convexity to relate optimization error to squared distance from the minimizer.
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 / 7 minor

Summary. This paper studies the problem of minimizing the k-th order partial derivative f = ∂_j^k g of an unknown function g over a compact convex set Θ, given a budget of N noisy zero-order queries of g at sequentially chosen points. The authors assume g has Hölder regularity β+k (β≥2), f is α-strongly convex, and standard boundedness/Lipschitz conditions hold. They propose a kernel-based gradient estimator using paired queries along the j-th coordinate (kernel K_k) and each gradient direction (kernel K_1), and analyze a projected stochastic gradient algorithm. The main results are: (1) Theorem 5.1, an upper bound on the expected optimization error of order d^{(2β+k-1)/(β+k)} N^{-(β-1)/(β+k)}; and (2) Theorem 5.2, a matching minimax lower bound of order N^{-(β-1)/(β+k)} over all sequential algorithms, establishing optimality in N. The gap between upper and lower bounds is a dimension-dependent polynomial factor, which is left as an open question.

Significance. The paper addresses a novel problem at the interface of zero-order optimization and nonparametric statistics: optimizing a derivative of an unknown function observed only through noisy function values. This 'indirect observation' setting is a natural generalization of standard zero-order optimization (k=0) and connects to deconvolution and linear inverse problems. The matching minimax rates in N are the central contribution and are established via a clean, self-contained proof chain. The upper bound uses a two-kernel construction with explicit moment conditions (Eqs. 7-8), a Chung-type recursion handling two forcing terms (Lemma 6.1), and an explicit bias-variance tradeoff (Remark 5.1). The lower bound constructs explicit adversarial functions g± = q ± r h^{β+k} Φ(x₁/h), verifies all class properties (a)-(g), and uses a KL-divergence reduction with a parameter choice (Eq. 59) that makes the KL divergence constant. The rate N^{-(β-1)/(β+k)} transparently reflects the loss of k units of smoothness from indirect observation. The dimension gap is honestly acknowledged.

major comments (2)
  1. Section 5.2, Theorem 5.2: The lower bound is stated under the assumption that the noise variables ξ_t are i.i.d. Gaussian (line: 'assume the oracle model (1), where ξ_t's are i.i.d. Gaussian with mean 0 and variance σ²'). The upper bound (Theorem 5.1) only requires E[ξ_t²] ≤ σ² with no independence or zero-mean assumption. This asymmetry is not highlighted in the abstract or introduction, where the results are presented as 'matching.' While the Gaussian restriction for lower bounds is standard practice, a brief remark explaining that the lower bound uses Gaussian noise (and thus the matching applies under this noise model) would improve transparency. This is a presentation issue rather than a correctness concern, as the upper bound is valid under weaker conditions.
  2. Section 6.2, proof of Theorem 5.2, verification of property (a) for β=2: The Hölder regularity of q is verified by bounding the Taylor remainder of order ℓ* = k+1, yielding |R_q(x,z)| ≤ C_q α ‖z-x‖^{k+2} (Eq. 46 and surrounding text). This is correct since β+k = k+2 when β=2. However, the case β>2 is handled by noting that q is a polynomial of degree k+2 and its Taylor expansion of order ℓ* ≥ k+2 is exact, so q ∈ F^{β+k}(0). This is fine, but the text should clarify that when β>2 is not an integer, ℓ* = ⌊β+k⌋ could still equal k+2 (e.g., β=2.5 gives ℓ*=k+2), in which case the exactness argument applies, but when β≥3, ℓ*≥k+3 and the degree-k+2 polynomial is also exactly represented. The current phrasing 'Since β+k > k+2 so ℓ* ≥ k+2' is slightly imprecise: for 2<β<3, ℓ*=k+2 exactly, not strictly greater. This does not affect the conclusion but the inequality should be ℓ* ≥ k+2.
minor comments (7)
  1. Section 1, definition of class F'_{α,β,k}: The condition (g) states 'f attains its minimum on Θ at a point x* ∈ int(Θ).' This is used in the lower bound proof (Section 6.2) to ensure x*± ∈ int(Θ) via the inclusion [-2h₀, 2h₀]^d ⊂ int(Θ) (Eq. 41). The proof correctly verifies this, but the main text could note that condition (g) is essential for the lower bound construction, not just convenience.
  2. Section 4.2, Eq. (6): The query points are written as x_t + hU_{t,l}e_j ± hV_{t,l}e_l. When l=j, both perturbations are along the same coordinate. The estimator (Eq. 9) still applies, but the analysis of Lemma 5.1 for l=j uses the bound |u±v| ≤ |u|+|v|, which is noted in the proof. This is handled correctly but a brief remark in Section 4.2 that the l=j case is non-degenerate would help the reader.
  3. Section 6.1, proof of Lemma 5.2, Eq. (18): The bound (g(x+s)-g(x-s))² ≤ 3L̄²h⁴/32 + 3G_g²h² uses ‖s‖ ≤ h/2. The factor 3/32 comes from 3·(L̄²/2)·(h/2)⁴·2 = 3L̄²h⁴/32. The intermediate step showing this substitution explicitly would aid verification, though the arithmetic is correct.
  4. Section 6.1, Lemma 6.1: The shift inequality (t+1+t₀)^q ≤ 2(t+t₀)^q is stated to follow from t₀ ≥ 2q. The proof uses (1+1/(2q))^q ≤ e^{1/2} ≤ 2. This is correct, but the bound e^{1/2} ≈ 1.649 < 2 is loose; this is fine for the result but could be tightened if sharper dimension dependence were sought.
  5. Section 7 (Discussion): The open question about the dimension exponent (2β+k-1)/(β+k) is well-motivated. The remark that 'standard Assouad-type constructions are not directly available' due to the interaction between Hölder regularity of g and strong convexity of ∂_j^k g is insightful and correctly identifies the technical obstacle.
  6. References: The paper by Wieczorek [2010] is cited as the only prior work on mode estimation under indirect observations. The connection to this paper is clearly delineated (sequential vs. i.i.d. observations). The reference list is appropriate and comprehensive.
  7. Notation: In Section 4.1, the Hölder class F_β(L) is defined with ℓ = ⌊β⌋ as 'the largest integer strictly less than β.' This is the standard convention (floor for non-integer β, β-1 for integer β). This should perhaps be stated more explicitly to avoid confusion with the convention ⌊β⌋ = β for integer β.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; derivation is self-contained and parameter-free

full rationale

The paper's two main results are derived from first principles without circular dependencies. The upper bound (Theorem 5.1) follows from: (1) a kernel estimator defined via explicit moment conditions (7)-(8), (2) a bias bound (Lemma 5.1) obtained by direct Hölder expansion and kernel moment cancellation, (3) a second-moment bound (Lemma 5.2) using the Lipschitz property of ∇g and noise variance, and (4) a Chung-type recursion (Lemma 6.1) proved from scratch. The rate N^{-(β-1)/(β+k)} emerges from balancing the variance term d·h^{-2(k+1)}/T against the bias term d·h^{2(β-1)}, yielding h ~ T^{-1/(2(β+k))}. No parameter is fitted to the target rate. The lower bound (Theorem 5.2) constructs explicit adversarial functions g± = q ± r·h^{β+k}·Φ(x₁/h), verifies all class properties (a)-(g), bounds the KL divergence via the chain rule, and derives minimizer separation via the implicit function theorem. The choice r_N = c*·N^{-1/2}·h_N^{-(β+k)} makes the KL divergence constant (not fitted to the rate), and the N^{-(β-1)/(β+k)} rate emerges from r²_N·h^{2(β-1)}_N = c*²·N^{-1}·h_N^{-2(k+1)} = c*²·N^{-(β-1)/(β+k)}. Self-citations to Akhavan et al. [2020, 2021, 2024a,b] are for context on the k=0 case and are not load-bearing. The textbook reference to Tsybakov [2009] for kernel construction is standard and external. The derivation chain is fully self-contained against external benchmarks.

Assumptions & free parameters 4 free parameters · 10 assumptions · 0 invented entities

The paper introduces no new physical entities, particles, forces, or postulated objects. The kernel-based gradient estimator (equation 9) is a new algorithmic construction, not an invented entity. The adversarial functions g± in the lower bound are explicit mathematical constructions, not postulated objects. All axioms are either standard mathematical results or domain assumptions about function regularity and noise that are clearly stated and standard in the nonparametric statistics literature.

free parameters (4)
  • γ (step size scale) = γ ≥ 4/α
    Algorithm parameter; lower bound chosen to satisfy q = αγ/2 ≥ 2 for Lemma 6.1. Any value satisfying the constraint works; the rate is independent of the specific choice.
  • t₀ (step size offset) = t₀ ≥ αγ
    Algorithm parameter; lower bound chosen to satisfy t₀ ≥ 2q for Lemma 6.1. Rate-independent.
  • κ (smoothing parameter scale) = κ > 0
    Controls h = min(κT^{-1/(2(β+k))}, 1). Any positive κ works; affects only constants, not the rate.
  • c* (lower bound construction constant) = min of four constraints
    Defined in equation (64) as the minimum of four upper bounds ensuring Hölder regularity, strong convexity, implicit function theorem validity, and KL divergence ≤ 1/8. Not fitted to data; chosen to satisfy structural constraints.
assumptions (10)
  • domain assumption g ∈ F^{β+k}(L): g has Hölder regularity β+k
    Section 1, property (a) of class F'_{α,β,k}. Required for the Taylor expansion in Lemma 5.1 and the bias bound of order h^{β-1}.
  • domain assumption f = ∂_j^k g is α-strongly convex on Θ
    Section 1, property (b). Load-bearing for the recursion (30) via inequality (28) and for the lower bound conversion (67).
  • domain assumption ∇f is L_f-Lipschitz on Θ
    Section 1, property (c). Used in the smoothness inequality (4) to convert distance bound to optimization error.
  • domain assumption ∥∇f(x)∥ ≤ G on Θ
    Section 1, property (d). Used in Lemma 5.2 to bound the squared conditional mean of the gradient estimator.
  • domain assumption ∥∇g(x)∥ ≤ G_g on Θ⁺
    Section 1, property (e). Used in Lemma 5.2, equation (18), to bound |g(x+s) - g(x-s)|.
  • domain assumption ∇g is L̄-Lipschitz on Θ⁺
    Section 1, property (f). Used in Lemma 5.2 to bound the remainder terms P and Q in the decomposition of g(x+s) - g(x-s).
  • domain assumption f attains its minimum at x* ∈ int(Θ)
    Section 1, property (g). Ensures ∇f(x*) = 0 and that the projection does not interfere with convergence.
  • domain assumption Noise variables ε_t have E[ε_t²] ≤ σ² (upper bound); ε_t i.i.d. Gaussian N(0,σ²) (lower bound)
    Section 2, oracle model (1). Upper bound requires only finite second moment; lower bound requires Gaussian noise for KL computation.
  • standard math Schwarz's theorem: mixed partial derivatives of order k+1 of g commute
    Equation (3), Section 4.1. Standard result; invoked to identify ∂_l ∂_j^k g = ∂_j^k ∂_l g.
  • standard math Chung-type recursion (Lemma 6.1)
    Section 6.1. Non-asymptotic version of a classical result (Chung 1954). Proved in the paper; closely related formulations in Moulines-Bach (2011) and Jiang et al. (2024).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Gradient-free stochastic optimization of derivatives under strong convexity." pith.science (2026). https://pith.science/paper/DTSFHTTI

@misc{pith2026260707249,
  author       = {Pith},
  title        = {Pith review of: Gradient-free stochastic optimization of derivatives under strong convexity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DTSFHTTI}},
  note         = {Machine review of arXiv:2607.07249}
}
abstract

We consider the problem of minimizing the $k$-th order partial derivative $f=\partial_j^k g$ of an unknown function $g$ along a fixed coordinate direction $j$, based on noisy queries of $g$. Assuming that $g$ has H\"older regularity ${\beta+k}$ for some $\beta\ge 2$, that $f$ is strongly convex on a compact convex set $\Theta\subset\mathbb{R}^d$ and that $g$ and $f$ satisfy mild boundedness and Lipschitz regularity conditions on $\Theta$, we propose a kernel-based estimator of $\nabla f$ and analyze the projected stochastic gradient algorithm driven by this estimator. We obtain a non-asymptotic upper bound on the optimization error of the order $d^{(2\beta+k-1)/(\beta+k)}\,N^{-(\beta-1)/(\beta+k)}$, where $N$ is the total number of queries. We also establish a minimax lower bound of the order $N^{-(\beta-1)/(\beta+k)}$ showing that this rate is optimal in $N$ over all sequential algorithms.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

300 extracted references · 300 canonical work pages

  1. [1]

    About adaptive coding on countable alphabets:

    Boucheron, St. About adaptive coding on countable alphabets:. IEEE Transactions on Information Theory , number =

    work page

  2. [2]

    About adaptive coding on countable alphabets , volume =

    Bontemps, Dominique and Boucheron, St. About adaptive coding on countable alphabets , volume =. IEEE Transactions on Information Theory , keywords =

    work page

  3. [3]

    Sparse adaptive

    Hutter, Marcus , booktitle =. Sparse adaptive

    work page

  4. [4]

    A hierarchical

    Teh, Yee Whye , booktitle =. A hierarchical

    work page

  5. [5]

    and Dahleh, Munther A

    Ohannessian, Mesrob I. and Dahleh, Munther A. , booktitle =. Rare probability estimation under regularly varying heavy tails , year =

    work page

  6. [6]

    On the concentration of the missing mass , volume =

    Berend, Daniel and Kontorovich, Aryeh , date-added =. On the concentration of the missing mass , volume =. Electronic Communications in Probability , keywords =

    work page

  7. [7]

    Revisiting Concentration of Missing Mass

    Skorski, Maciej , date-added =. Revisiting concentration of missing mass , year =. arXiv preprint arXiv:2005.10018 , keywords =

    work page Pith review arXiv 2005

  8. [8]

    Convergence guarantees for the

    Painsky, Amichai , date-added =. Convergence guarantees for the. Journal of Machine Learning Research , keywords =

    work page

Show all 300 references
  1. [9]

    Concentration bounds for unigram language models , volume =

    Drukh, Evgeny and Mansour, Yishay , date-added =. Concentration bounds for unigram language models , volume =. Journal of Machine Learning Research , keywords =

  2. [10]

    , booktitle =

    McAllester, David and Schapire, Robert E. , booktitle =. On the convergence rate of

  3. [11]

    Concentration inequalities for the missing mass and for histogram rule error , volume =

    McAllester, David and Ortiz, Luis , date-added =. Concentration inequalities for the missing mass and for histogram rule error , volume =. Journal of Machine Learning Research , keywords =

  4. [12]

    and Goodman, Joshua , date-added =

    Chen, Stanley F. and Goodman, Joshua , date-added =. An empirical study of smoothing techniques for language modeling , volume =. Computer Speech & Language , keywords =

  5. [13]

    Local risk bounds for statistical aggregation , year =

    Mourtada, Jaouad and Va. Local risk bounds for statistical aggregation , year =

  6. [14]

    Optimal exponential bounds for aggregation of estimators for the

    Butucea, Cristina and Delmas, Jean-Fran. Optimal exponential bounds for aggregation of estimators for the. Electronic Journal of Statistics , keywords =

  7. [15]

    Improved backing-off for

    Kneser, Reinhard and Ney, Hermann , booktitle =. Improved backing-off for

  8. [16]

    On structuring probabilistic dependences in stochastic language modelling , volume =

    Ney, Hermann and Essen, Ute and Kneser, Reinhard , date-added =. On structuring probabilistic dependences in stochastic language modelling , volume =. Computer Speech & Language , keywords =

  9. [17]

    and Orlitsky, Alon and Pichapati, Venkatadheeraj , booktitle =

    Falahatgar, Moein and Ohannessian, Mesrob I. and Orlitsky, Alon and Pichapati, Venkatadheeraj , booktitle =. The power of absolute discounting: all-dimensional distribution estimation , year =

  10. [18]

    Doubly-competitive distribution estimation , year =

    Hao, Yi and Orlitsky, Alon , booktitle =. Doubly-competitive distribution estimation , year =

  11. [19]

    Optimal probability estimation with applications to prediction and classification , year =

    Acharya, Jayadev and Jafarpour, Ashkan and Orlitsky, Alon and Suresh, Ananda Theertha , booktitle =. Optimal probability estimation with applications to prediction and classification , year =

  12. [20]

    Competitive distribution estimation:

    Orlitsky, Alon and Suresh, Ananda Theertha , booktitle =. Competitive distribution estimation:

  13. [21]

    Richard and Richardson, Thomas S

    Guo, F. Richard and Richardson, Thomas S. , journal =. Chernoff-type concentration of empirical probabilities in relative entropy , volume =

  14. [22]

    Concentration bounds for discrete distribution estimation in

    Canonne, Cl. Concentration bounds for discrete distribution estimation in. Proceedings of the

  15. [23]

    Optimal prediction of

    Han, Yanjun and Jana, Soham and Wu, Yihong , date-added =. Optimal prediction of. IEEE Transactions on Information Theory , keywords =

  16. [24]

    Optimal prediction of

    Han, Yanjun and Jana, Soham and Wu, Yihong , date-added =. Optimal prediction of. Advances in Neural Information Processing Systems , keywords =

  17. [25]

    Bhattacharyya, Arnab and Gayen, Sutanu and Price, Eric and Vinodchandran, N. V. , booktitle =. Near-optimal learning of tree-structured distributions by

  18. [26]

    Bernstein polynomials and learning theory , volume =

    Braess, Dietrich and Sauer, Thomas , date-added =. Bernstein polynomials and learning theory , volume =. Journal of Approximation Theory , keywords =

  19. [27]

    The performance of universal encoding , volume =

    Krichevsky, Raphail and Trofimov, Victor , date-added =. The performance of universal encoding , volume =. IEEE Transactions on Information Theory , keywords =

  20. [28]

    Sharp concentration inequalities for the centred relative entropy , volume =

    Bhatt, Alankrita and Pensia, Ankit , date-added =. Sharp concentration inequalities for the centred relative entropy , volume =. Information and Inference: A Journal of the IMA , keywords =

  21. [29]

    Concentration inequalities for the empirical distribution of discrete distributions: beyond the method of types , volume =

    Mardia, Jay and Jiao, Jiantao and T. Concentration inequalities for the empirical distribution of discrete distributions: beyond the method of types , volume =. Information and Inference: A Journal of the IMA , keywords =

  22. [30]

    Optimal estimation of high-order missing masses, and the rare-type match problem , year =

    Favaro, Stefano and Naulet, Zacharie , date-added =. Optimal estimation of high-order missing masses, and the rare-type match problem , year =. Preprint arXiv:2306.14998 , keywords =

    arXiv

  23. [31]

    A short note on learning discrete distributions , year =

    Canonne, Cl. A short note on learning discrete distributions , year =. arXiv preprint arXiv:2002.11457 , keywords =

    2002 arXiv

  24. [32]

    Probability and computing:

    Mitzenmacher, Michael and Upfal, Eli , date-added =. Probability and computing:

  25. [33]

    Negative association of random variables with applications , volume =

    Joag-Dev, Kumar and Proschan, Frank , date-added =. Negative association of random variables with applications , volume =. The Annals of Statistics , keywords =

  26. [34]

    Generalized multi-view model:

    Chhor, Julien and Klopp, Olga and Tsybakov, Alexandre , journal =. Generalized multi-view model:

  27. [35]

    About the non-asymptotic behaviour of

    Birg. About the non-asymptotic behaviour of. Journal of Statistical Planning and Inference , keywords =

  28. [36]

    From robust tests to

    Baraud, Yannick , journal =. From robust tests to

  29. [37]

    Active regression via linear-sample sparsification , volume =

    Chen, Xue and Price, Eric , booktitle =. Active regression via linear-sample sparsification , volume =

  30. [38]

    Dimension-free bounds for sums of independent matrices and simple tensors via the variational principle , volume =

    Zhivotovskiy, Nikita , journal =. Dimension-free bounds for sums of independent matrices and simple tensors via the variational principle , volume =

  31. [39]

    On a conjecture of

    Park, Jinyoung and Pham, Huy Tuan , journal =. On a conjecture of

  32. [40]

    Finite-sample performance of the maximum likelihood estimator in logistic regression , year =

    Chardon, Hugo and Lerasle, Matthieu and Mourtada, Jaouad , journal =. Finite-sample performance of the maximum likelihood estimator in logistic regression , year =

  33. [41]

    Probability in high dimension , year =

    Van Handel, Ramon , journal =. Probability in high dimension , year =

  34. [42]

    , journal =

    Sudakov, Vladimir N. , journal =. Typical distributions of linear functionals in finite-dimensional spaces of higher dimension , volume =

  35. [43]

    Concentration and

    Bobkov, Sergey and Chistyakov, Gennadiy and G. Concentration and

  36. [44]

    An Introduction to Probability Theory and Its Applications , volume =

    Feller, William , edition =. An Introduction to Probability Theory and Its Applications , volume =

  37. [45]

    , booktitle =

    Bobkov, Sergey G. , booktitle =. Edgeworth corrections in randomized central limit theorems , volume =

  38. [46]

    An introduction to the

    Naor, Assaf , journal =. An introduction to the

  39. [47]

    Conjugate priors for exponential families , volume =

    Diaconis, Persi and Ylvisaker, Donald , journal =. Conjugate priors for exponential families , volume =

  40. [48]

    Robust estimation of a regression function in exponential families , volume =

    Baraud, Yannick and Chen, Juntong , journal =. Robust estimation of a regression function in exponential families , volume =

  41. [49]

    , booktitle =

    Tan, Kai and Bellec, Pierre C. , booktitle =. Multinomial logistic regression: Asymptotic normality on null covariates in high-dimensions , volume =

  42. [50]

    , journal =

    Zhao, Qian and Sur, Pragya and Candes, Emmanuel J. , journal =. The asymptotic distribution of the

  43. [51]

    Sharp asymptotics and optimal performance for inference in binary models , year =

    Taheri, Hossein and Pedarsani, Ramtin and Thrampoulidis, Christos , booktitle =. Sharp asymptotics and optimal performance for inference in binary models , year =

  44. [52]

    Estimation bounds and sharp oracle inequalities of regularized procedures with

    Alquier, Pierre and Cottet, Vincent and Lecu. Estimation bounds and sharp oracle inequalities of regularized procedures with. The Annals of Statistics , keywords =

  45. [53]

    and Liu, Richard C

    Donoho, David L. and Liu, Richard C. , journal =. Geometrizing rates of convergence,

  46. [54]

    Mean field asymptotics in high-dimensional statistics: From exact results to efficient algorithms , year =

    Montanari, Andrea , booktitle =. Mean field asymptotics in high-dimensional statistics: From exact results to efficient algorithms , year =

  47. [55]

    High-probability risk bounds via sequential predictors , year =

    van der Hoeven, Dirk and Zhivotovskiy, Nikita and Cesa-Bianchi, Nicol. High-probability risk bounds via sequential predictors , year =. Preprint arXiv:2308.07588 , keywords =

    arXiv

  48. [56]

    Localization, convexity, and star aggregation , year =

    Vijaykumar, Suhas , booktitle =. Localization, convexity, and star aggregation , year =

  49. [57]

    Tight lower bounds on the

    Csik. Tight lower bounds on the. Journal of Machine Learning Research , keywords =

  50. [58]

    Variations on the

    Klartag, Bo'az and Sodin, Sasha , journal =. Variations on the

  51. [59]

    Logarithmic bounds for isoperimetry and slices of convex sets , year =

    Klartag, Bo'az , journal =. Logarithmic bounds for isoperimetry and slices of convex sets , year =

  52. [60]

    The central limit theorem around 1935 , volume =

    Le Cam, Lucien , journal =. The central limit theorem around 1935 , volume =

    1935

  53. [61]

    , booktitle =

    Shevtsova, Irina G. , booktitle =. An improvement of convergence rate estimates in the

  54. [62]

    Tyurin, I. S. , journal =. A refinement of the remainder in the

  55. [63]

    , journal =

    Dudley, Richard M. , journal =. Central limit theorems for empirical measures , volume =

  56. [64]

    and Chervonenkis, Alexey Ya

    Vapnik, Vladimir N. and Chervonenkis, Alexey Ya. , journal =. On the uniform convergence of relative frequencies of events to their probabilities , volume =

  57. [65]

    Essai philosophique sur les probabilit

    Laplace, Pierre-Simon de , edition =. Essai philosophique sur les probabilit

  58. [66]

    Sphere packing numbers for subsets of the

    Haussler, David , journal =. Sphere packing numbers for subsets of the

  59. [67]

    and Shepp, Lawrence A

    Landau, Henry J. and Shepp, Lawrence A. , journal =. On the supremum of a

  60. [68]

    Tomczak-Jaegermann, Nicole , fjournal =. Dualit. C. R. Acad. Sci., Paris, S

  61. [69]

    Learning Theory from First Principles , year =

    Bach, Francis , keywords =. Learning Theory from First Principles , year =

  62. [70]

    Concentration inequalities in the infinite urn scheme for occupancy counts and the missing mass, with applications , volume =

    Ben-Hamou, Anna and Boucheron, St. Concentration inequalities in the infinite urn scheme for occupancy counts and the missing mass, with applications , volume =. Bernoulli , keywords =

  63. [71]

    Integral geometry for the 1-norm , volume =

    Leinster, Tom , journal =. Integral geometry for the 1-norm , volume =

  64. [72]

    Empirical processes with a bounded _1 diameter , volume =

    Mendelson, Shahar , journal =. Empirical processes with a bounded _1 diameter , volume =

  65. [73]

    Reconstruction and subgaussian operators in asymptotic geometric analysis , volume =

    Mendelson, Shahar and Pajor, Alain and Tomczak-Jaegermann, Nicole , journal =. Reconstruction and subgaussian operators in asymptotic geometric analysis , volume =

  66. [74]

    Talagrand, Michel , journal =. New

  67. [75]

    and Linde, Werner , journal =

    Li, Wenbo V. and Linde, Werner , journal =. Approximation, metric entropy and small ball estimates for

  68. [76]

    , journal =

    Kuelbs, James and Li, Wenbo V. , journal =. Metric entropy and the small ball problem for

  69. [77]

    and Shao, Qi-Man , journal =

    Li, Wenbo V. and Shao, Qi-Man , journal =. Gaussian processes:

  70. [78]

    and Montgomery-Smith, Stephen J

    Dilworth, Stephen J. and Montgomery-Smith, Stephen J. , journal =. The distribution of vector-valued

  71. [79]

    , journal =

    Montgomery-Smith, Stephen J. , journal =. The distribution of

  72. [80]

    Probability inequalities for the sum of independent random variables , volume =

    Bennett, George , journal =. Probability inequalities for the sum of independent random variables , volume =

  73. [81]

    Statistical behavior and consistency of classification methods based on convex risk minimization , volume =

    Zhang, Tong , journal =. Statistical behavior and consistency of classification methods based on convex risk minimization , volume =

  74. [82]

    and Nemirovski, Arkadi S

    Juditsky, Anatoli B. and Nemirovski, Arkadi S. , journal =. Nonparametric estimation by convex programming , volume =

  75. [83]

    On estimation of a probability density function and mode , volume =

    Parzen, Emanuel , journal =. On estimation of a probability density function and mode , volume =

  76. [84]

    Estimation of the mode , volume =

    Chernoff, Herman , journal =. Estimation of the mode , volume =

  77. [85]

    and Offord, Albert C

    Littlewood, John E. and Offord, Albert C. , journal =. On the number of real roots of a random algebraic equation

  78. [86]

    On a lemma of

    Erd. On a lemma of. Bulletin of the American Mathematical Society , keywords =

  79. [87]

    , keywords =

    Petrov, Valentin V. , keywords =. Limit theorems of probability theory

  80. [88]

    On the dependence of the

    Bentkus, Vidmantas , journal =. On the dependence of the

  81. [89]

    A multivariate

    Rai. A multivariate. Bernoulli , keywords =

  82. [90]

    Chen, Louis H. Y. and Shao, Qi-Man , journal =. A non-uniform

  83. [91]

    Feller, William , journal =. On the

  84. [92]

    , journal =

    Rogozin, Boris A. , journal =. An estimate for concentration functions , volume =

  85. [93]

    A sharper form of the

    Kesten, Harry , journal =. A sharper form of the

  86. [94]

    Zonoids and generalisations , year =

    Goodey, Paul and Weil, Wolfgang , booktitle =. Zonoids and generalisations , year =

  87. [95]

    High-dimensional analysis of double descent for linear regression with random projections , year =

    Bach, Francis , journal =. High-dimensional analysis of double descent for linear regression with random projections , year =

  88. [96]

    Robust machine learning by median-of-means: theory and practice , volume =

    Lecu. Robust machine learning by median-of-means: theory and practice , volume =. The Annals of Statistics , keywords =

  89. [97]

    , journal =

    Hastie, Trevor and Montanari, Andrea and Rosset, Saharon and Tibshirani, Ryan J. , journal =. Surprises in high-dimensional ridgeless least squares interpolation , volume =

  90. [98]

    On the optimal weighted _2 regularization in overparameterized linear regression , volume =

    Wu, Denny and Xu, Ji , journal =. On the optimal weighted _2 regularization in overparameterized linear regression , volume =

  91. [99]

    A new isoperimetric inequality and the concentration of measure phenomenon , volume =

    Talagrand, Michel , booktitle =. A new isoperimetric inequality and the concentration of measure phenomenon , volume =

  92. [100]

    , booktitle =

    Szarek, Stanislaw J. , booktitle =. Convexity, complexity, and high dimensions , volume =

  93. [101]

    Sudakov minoration principle and supremum of some processes , volume =

    Lata. Sudakov minoration principle and supremum of some processes , volume =. Geometric and Functional Analysis , keywords =

  94. [102]

    The supremum of some canonical processes , volume =

    Talagrand, Michel , journal =. The supremum of some canonical processes , volume =

  95. [103]

    A short proof, based on mixed volumes, of

    Gurvits, Leonid , journal =. A short proof, based on mixed volumes, of

  96. [104]

    , journal =

    Shephard, Geoffrey C. , journal =. Inequalities between mixed volumes of convex sets , volume =

  97. [105]

    Universal coding, intrinsic volumes, and metric complexity , year =

    Mourtada, Jaouad , journal =. Universal coding, intrinsic volumes, and metric complexity , year =

  98. [106]

    Some inequalities for

    Gordon, Yehoram , journal =. Some inequalities for

  99. [107]

    A short proof of concentration of the intrinsic volume sequence of a convex body , year =

    Mourtada, Jaouad , journal =. A short proof of concentration of the intrinsic volume sequence of a convex body , year =

  100. [108]

    , journal =

    Davisson, Lee D. , journal =. Universal noiseless coding , volume =

  101. [109]

    , journal =

    Kolmogorov, Andrey N. , journal =. Three approaches to the quantitative definition of information , volume =

  102. [110]

    Various thresholds for _1 -optimization in compressed sensing , year =

    Stojnic, Mihailo , journal =. Various thresholds for _1 -optimization in compressed sensing , year =

  103. [111]

    Near-optimal signal recovery from random projections:

    Cand. Near-optimal signal recovery from random projections:. IEEE Transactions on Information Theory , keywords =

  104. [112]

    , journal =

    Donoho, David L. , journal =. Compressed sensing , volume =

  105. [113]

    On sparse reconstruction from

    Rudelson, Mark and Vershynin, Roman , journal =. On sparse reconstruction from

  106. [114]

    and Willsky, Alan S

    Chandrasekaran, Venkat and Recht, Benjamin and Parrilo, Pablo A. and Willsky, Alan S. , journal =. The convex geometry of linear inverse problems , volume =

  107. [115]

    , journal =

    Milman, Vitali D. , journal =. New proof of the theorem of

  108. [116]

    Small ball probability estimates in terms of width , volume =

    Lata. Small ball probability estimates in terms of width , volume =. Studia Mathematica , keywords =

  109. [117]

    On a quantitative reversal of

    Paouris, Grigoris and Pivovarov, Peter and Valettas, Petros , journal =. On a quantitative reversal of

  110. [118]

    Small ball probability and

    Klartag, Bo'az and Vershynin, Roman , journal =. Small ball probability and

  111. [119]

    Cordero-Erausquin, Dario and Fradelizi, Matthieu and Maurey, Bernard , journal =. The

  112. [120]

    Alquier, Pierre , journal =

  113. [121]

    , booktitle =

    Cover, Thomas M. , booktitle =. Behavior of sequential predictors of binary sequences , year =

  114. [122]

    , journal =

    Kolmogorov, Andrey N. , journal =. The method of the median in the theory of errors , volume =

  115. [123]

    Estimating location parameters in sample-heterogeneous distributions , volume =

    Pensia, Ankit and Jog, Varun and Loh, Po-Ling , journal =. Estimating location parameters in sample-heterogeneous distributions , volume =

  116. [124]

    Learning entangled single-sample

    Chierichetti, Flavio and Dasgupta, Anirban and Kumar, Ravi and Lattanzi, Silvio , booktitle =. Learning entangled single-sample

  117. [125]

    , journal =

    Mizera, Ivan and Wellner, Jon A. , journal =. Necessary and sufficient conditions for weak consistency of the median of independent but not identically distributed random variables , volume =

  118. [126]

    Learning entangled single-sample

    Liang, Yingyu and Yuan, Hui , booktitle =. Learning entangled single-sample

  119. [127]

    Non-asymptotic bounds for percentiles of independent non-identical random variables , volume =

    Xia, Dong , journal =. Non-asymptotic bounds for percentiles of independent non-identical random variables , volume =

  120. [128]

    On mean estimation for heteroscedastic random variables , volume =

    Devroye, Luc and Lattanzi, Silvio and Lugosi, G. On mean estimation for heteroscedastic random variables , volume =. Annales de l'Institut Henri Poincar

  121. [129]

    Learning subgaussian classes: Upper and minimax bounds , year =

    Lecu. Learning subgaussian classes: Upper and minimax bounds , year =

  122. [130]

    Tight bounds on minimax regret under logarithmic loss via self-concordance , year =

    Bilodeau, Blair and Foster, Dylan and Roy, Daniel , booktitle =. Tight bounds on minimax regret under logarithmic loss via self-concordance , year =

  123. [131]

    Wavelets, Approximation, and Statistical Applications , volume =

    H. Wavelets, Approximation, and Statistical Applications , volume =

  124. [132]

    Estimating a regression function , volume =

    van de Geer, Sara , journal =. Estimating a regression function , volume =

  125. [133]

    A Mathematical Introduction to Compressive Sensing , year =

    Foucart, Simon and Rauhut, Holger , keywords =. A Mathematical Introduction to Compressive Sensing , year =

  126. [134]

    and Johnstone, Iain M

    Donoho, David L. and Johnstone, Iain M. , journal =. Minimax risk over

  127. [135]

    Adaptive minimax regret against smooth logarithmic losses over high-dimensional l1-balls via envelope complexity , volume =

    Miyaguchi, Kohei and Yamanishi, Kenji , booktitle =. Adaptive minimax regret against smooth logarithmic losses over high-dimensional l1-balls via envelope complexity , volume =

  128. [136]

    Entropy numbers of diagonal operators between symmetric

    Sch. Entropy numbers of diagonal operators between symmetric. Journal of Approximation Theory , keywords =

  129. [137]

    van Erven, Tim and Harremo. R. IEEE Transactions on Information Theory , keywords =

  130. [138]

    Mutual information, metric entropy and cumulative relative entropy risk , volume =

    Haussler, David and Opper, Manfred , journal =. Mutual information, metric entropy and cumulative relative entropy risk , volume =

  131. [139]

    On measures of entropy and information , volume =

    R. On measures of entropy and information , volume =. Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability , keywords =

  132. [140]

    Information Theory and Statistics: A Tutorial , volume =

    Csisz. Information Theory and Statistics: A Tutorial , volume =. Foundations and Trends in Communications and Information Theory , keywords =

  133. [141]

    and Vetterli, Martin and DeVore, Ronald A

    Donoho, David L. and Vetterli, Martin and DeVore, Ronald A. and Daubechies, Ingrid , journal =. Data compression and harmonic analysis , volume =

  134. [142]

    Entropy, compactness and the approximation of operators , year =

    Carl, Bernd and Stephani, Irmtraud , keywords =. Entropy, compactness and the approximation of operators , year =

  135. [143]

    and Mendelson, Shahar and Pajor, Alain , journal =

    Gordon, Yehoram and Litvak, Alexander E. and Mendelson, Shahar and Pajor, Alain , journal =. Gaussian averages of interpolated bodies and applications to approximate reconstruction , volume =

  136. [144]

    Information Theory: From Coding to Learning , year =

    Polyanskiy, Yury and Wu, Yihong , keywords =. Information Theory: From Coding to Learning , year =

  137. [145]

    The total variation distance between high-dimensional

    Devroye, Luc and Mehrabian, Abbas and Reddad, Tommy , journal =. The total variation distance between high-dimensional

  138. [146]

    Yu, Bin , booktitle =

  139. [147]

    , journal =

    Klartag, Boaz and Milman, Vitali D. , journal =. Geometry of log-concave functions and measures , volume =

  140. [148]

    Stochastic geometry to generalize the

    O'Reilly, Eliza and Tran, Ngoc Mai , journal =. Stochastic geometry to generalize the

  141. [149]

    Concentration inequalities for ultra log-concave distributions , volume =

    Aravinda, Heshan and Marsiglietti, Arnaud and Melbourne, James , journal =. Concentration inequalities for ultra log-concave distributions , volume =

  142. [150]

    On prediction of individual sequences , volume =

    Cesa-Bianchi, Nicolo and Lugosi, G. On prediction of individual sequences , volume =. The Annals of Statistics , keywords =

  143. [151]

    , journal =

    Costa, Max and Cover, Thomas M. , journal =. On the similarity of the entropy power inequality and the

  144. [152]

    and Thomas, Joy A

    Dembo, Amir and Cover, Thomas M. and Thomas, Joy A. , journal =. Information theoretic inequalities , volume =

  145. [153]

    Mean Field Models for Spin Glasses

    Talagrand, Michel , keywords =. Mean Field Models for Spin Glasses. Volume I: Basic Examples , volume =

  146. [154]

    Algorithms in Real Algebraic Geometry , year =

    Basu, Saugata and Pollack, Richard and Roy, Marie-Fran. Algorithms in Real Algebraic Geometry , year =

  147. [155]

    Isoperimetry and integrability of the sum of independent Banach-space valued random variables , volume =

    Talagrand, Michel , journal =. Isoperimetry and integrability of the sum of independent Banach-space valued random variables , volume =

  148. [156]

    Universal coding, information, prediction, and estimation , volume =

    Rissanen, Jorma , journal =. Universal coding, information, prediction, and estimation , volume =

  149. [157]

    Bobkov, Sergey and Ledoux, Michel , journal =

  150. [158]

    Martingales with values in uniformly convex spaces , volume =

    Pisier, Gilles , journal =. Martingales with values in uniformly convex spaces , volume =

  151. [159]

    and Schechtman, Gideon and Zinn, Joel , journal =

    Johnson, William B. and Schechtman, Gideon and Zinn, Joel , journal =. Best constants in moment inequalities for linear combinations of independent and exchangeable random variables , volume =

  152. [160]

    Kahane, Jean-Pierre , journal =. Une in

  153. [161]

    Sur le chaos multiplicatif , volume =

    Kahane, Jean-Pierre , journal =. Sur le chaos multiplicatif , volume =

  154. [162]

    , keywords =

    Bennett, Colin and Sharpley, Robert C. , keywords =. Interpolation of Operators , year =

  155. [163]

    Classical

    Grafakos, Loukas , edition =. Classical

  156. [164]

    Upper and lower bounds for stochastic processes: Decomposition Theorems , year =

    Talagrand, Michel , edition =. Upper and lower bounds for stochastic processes: Decomposition Theorems , year =

  157. [165]

    Tail bounds via generic chaining , volume =

    Dirksen, Sjoerd , journal =. Tail bounds via generic chaining , volume =

  158. [166]

    A tail inequality for suprema of unbounded empirical processes with applications to

    Adamczak, Rados. A tail inequality for suprema of unbounded empirical processes with applications to. Electronic Journal of Probability , keywords =

  159. [167]

    A simple tool for bounding the deviation of random matrices on geometric sets , volume =

    Liaw, Christopher and Mehrabian, Abbas and Plan, Yaniv and Vershynin, Roman , booktitle =. A simple tool for bounding the deviation of random matrices on geometric sets , volume =

  160. [168]

    Sub-Gaussian matrices on sets:

    Jeong, Halyun and Li, Xiaowei and Plan, Yaniv and Yilmaz, Ozgur , journal =. Sub-Gaussian matrices on sets:

  161. [169]

    , journal =

    Dvoretzky, Aryeh and Rogers, Claude A. , journal =. Absolute and unconditional convergence in normed linear spaces , volume =

  162. [170]

    Nearly optimal robust mean estimation via empirical characteristic function , volume =

    Bahmani, Sohail , journal =. Nearly optimal robust mean estimation via empirical characteristic function , volume =

  163. [171]

    An error bound in the

    Chatterjee, Sourav , journal =. An error bound in the

  164. [172]

    and Su, Francis E

    Gibbs, Alison L. and Su, Francis E. , journal =. On choosing and bounding probability metrics , volume =

  165. [173]

    Optimal sub-gaussian mean estimation in

    Lee, Jasper CH and Valiant, Paul , journal =. Optimal sub-gaussian mean estimation in

  166. [174]

    de Araujo, Aloisio P. and Gin. The central limit theorem for real and

  167. [175]

    , booktitle =

    Jain, Naresh C. , booktitle =. Central limit theorem and related questions in

  168. [176]

    and Snobar, M

    Kadets, Mikhail I. and Snobar, M. G. , journal =. Certain functionals on the

  169. [177]

    Pisier, Gilles , journal =. Le th

  170. [178]

    Finite-sample concentration of the empirical relative entropy around its mean , year =

    Agrawal, Rohit , journal =. Finite-sample concentration of the empirical relative entropy around its mean , year =

  171. [179]

    The method of types , volume =

    Csisz. The method of types , volume =. IEEE Transactions on Information Theory , keywords =

  172. [180]

    The convex hull of random points with spherically symmetric distributions , volume =

    Affentranger, F , journal =. The convex hull of random points with spherically symmetric distributions , volume =

  173. [181]

    Calka, Pierre and Yukich, J. E. , journal =. Variance asymptotics and scaling limits for

  174. [182]

    The convex hull of a random set of points , volume =

    Efron, Bradley , journal =. The convex hull of a random set of points , volume =

  175. [183]

    Expected volumes of

    Kabluchko, Zakhar and Zaporozhets, Dmitry , journal =. Expected volumes of

  176. [184]

    Minimum complexity density estimation , volume =

    Barron, Andrew R and Cover, Thomas M , journal =. Minimum complexity density estimation , volume =

  177. [185]

    and Lieb, Elliott H

    Brascamp, Herm J. and Lieb, Elliott H. , journal =. On extensions of the

  178. [186]

    and Lieb, Elliott H

    Brascamp, Herm J. and Lieb, Elliott H. , journal =. Best constants in

  179. [187]

    Universal Coding and Order Identification by Model Selection Methods , year =

    Gassiat,. Universal Coding and Order Identification by Model Selection Methods , year =

  180. [188]

    Transportation cost for

    Talagrand, Michel , journal =. Transportation cost for

  181. [189]

    A new perspective on robust M -estimation:

    Zhou, Wen-Xin and Bose, Koushiki and Fan, Jianqing and Liu, Han , journal =. A new perspective on robust M -estimation:

  182. [190]

    The influence of variables on

    Kahn, Jeff and Kalai, Gil and Linial, Nathan , booktitle =. The influence of variables on

  183. [191]

    Talagrand, Michel , journal =. On

  184. [192]

    A simple proof of the

    Royen, Thomas , journal =. A simple proof of the

  185. [193]

    Probabilistic methods in the geometry of

    Pisier, Gilles , booktitle =. Probabilistic methods in the geometry of

  186. [194]

    Aggregated estimators and empirical complexity for least square regression , volume =

    Audibert, Jean-Yves , journal =. Aggregated estimators and empirical complexity for least square regression , volume =

  187. [195]

    , keywords =

    Brown, Lawrence D. , keywords =. Fundamentals of statistical exponential families with applications in statistical decision theory , volume =

  188. [196]

    Estimation of moments of sums of independent real random variables , volume =

    Lata. Estimation of moments of sums of independent real random variables , volume =. The Annals of Probability , keywords =

  189. [197]

    An elementary analysis of ridge regression with random design , volume =

    Mourtada, Jaouad and Rosasco, Lorenzo , journal =. An elementary analysis of ridge regression with random design , volume =

  190. [198]

    On some inequalities for

    Lata. On some inequalities for. Proceedings of the ICM , keywords =

  191. [199]

    Introduction to Banach Spaces: Analysis and Probability: Volume 1 , year =

    Li, Daniel and Queff. Introduction to Banach Spaces: Analysis and Probability: Volume 1 , year =

  192. [200]

    Type, cotype and

    Maurey, Bernard , journal =. Type, cotype and

  193. [201]

    and Lindenstrauss, Joram , booktitle =

    Johnson, William B. and Lindenstrauss, Joram , booktitle =. Basic concepts in the geometry of

  194. [202]

    Gardner, Richard , journal =. The

  195. [203]

    Ehrhard, Antoine , journal =. Sym

  196. [204]

    Borell, Christer , journal =. The

  197. [205]

    Geometric and functional inequalities for log-concave probability sequences , year =

    Marsiglietti, Arnaud and Melbourne, James , journal =. Geometric and functional inequalities for log-concave probability sequences , year =

  198. [206]

    High-dimensional estimation with geometric constraints , volume =

    Plan, Yaniv and Vershynin, Roman and Yudovina, Elena , journal =. High-dimensional estimation with geometric constraints , volume =

  199. [207]

    , keywords =

    Artstein-Avidan, Shiri and Giannopoulos, Apostolos and Milman, Vitali D. , keywords =. Asymptotic Geometric Analysis, Part

  200. [208]

    The best constants in the

    Haagerup, Uffe , journal =. The best constants in the

  201. [209]

    Ambrose and Shephard, Geoffrey C

    Rogers, C. Ambrose and Shephard, Geoffrey C. , journal =. The difference body of a convex body , volume =

  202. [210]

    , journal =

    Tsigler, Alexander and Bartlett, Peter L. , journal =. Benign overfitting in ridge regression , volume =

  203. [211]

    Processus

    Chevet, Simone , journal =. Processus

  204. [212]

    Non commutative

    Lust-Piquard, Fran. Non commutative. Arkiv f

  205. [213]

    Introduction to Operator Space Theory , volume =

    Pisier, Gilles , keywords =. Introduction to Operator Space Theory , volume =

  206. [214]

    Computing mixed discriminants, mixed volumes, and permanents , volume =

    Barvinok, Alexander , journal =. Computing mixed discriminants, mixed volumes, and permanents , volume =

  207. [215]

    Regularity of

    Talagrand, Michel , journal =. Regularity of

  208. [216]

    An improper estimator with optimal excess risk in misspecified density estimation and logistic regression , volume =

    Mourtada, Jaouad and Ga. An improper estimator with optimal excess risk in misspecified density estimation and logistic regression , volume =. Journal of Machine Learning Research , keywords =

  209. [217]

    Contributions

    Mourtada, Jaouad , keywords =. Contributions

  210. [218]

    Paouris, Grigoris and Valettas, Petros and others , journal =. A

  211. [219]

    Dimension reduction by random hyperplane tessellations , volume =

    Plan, Yaniv and Vershynin, Roman , journal =. Dimension reduction by random hyperplane tessellations , volume =

  212. [220]

    From low probability to high confidence in stochastic convex optimization , volume =

    Davis, Damek and Drusvyatskiy, Dmitriy and Xiao, Lin and Zhang, Junyu , journal =. From low probability to high confidence in stochastic convex optimization , volume =

  213. [221]

    Concentration of measure and isoperimetric inequalities in product spaces , volume =

    Talagrand, Michel , journal =. Concentration of measure and isoperimetric inequalities in product spaces , volume =

  214. [222]

    Heat flow derivatives and minimum mean-square error in

    Ledoux, Michel , journal =. Heat flow derivatives and minimum mean-square error in

  215. [223]

    On the minimax rate of the

    Neykov, Matey , journal =. On the minimax rate of the

  216. [224]

    Nearly optimal minimax estimator for high-dimensional sparse linear regression , volume =

    Zhang, Li , journal =. Nearly optimal minimax estimator for high-dimensional sparse linear regression , volume =

  217. [225]

    Near-optimality of linear recovery from indirect observations , volume =

    Juditsky, Anatoli and Nemirovski, Arkadi , journal =. Near-optimality of linear recovery from indirect observations , volume =

  218. [226]

    Near-optimality of linear recovery in

    Juditsky, Anatoli and Nemirovski, Arkadi , journal =. Near-optimality of linear recovery in

  219. [227]

    and Schechtman, Gideon , keywords =

    Milman, Vitali D. and Schechtman, Gideon , keywords =. Asymptotic Theory of Finite Dimensional Normed Spaces , volume =

  220. [228]

    Chaining, interpolation and convexity

    van Handel, Ramon , journal =. Chaining, interpolation and convexity

  221. [229]

    Chaining, interpolation, and convexity , volume =

    Van Handel, Ramon , journal =. Chaining, interpolation, and convexity , volume =

  222. [230]

    Dimensional improvements of the logarithmic

    Bolley, Fran. Dimensional improvements of the logarithmic. The Annals of Probability , keywords =

  223. [231]

    The entropic barrier is n -self-concordant , year =

    Chewi, Sinho , journal =. The entropic barrier is n -self-concordant , year =

  224. [232]

    A universal procedure for aggregating estimators , volume =

    Goldenshluger, Alexander , journal =. A universal procedure for aggregating estimators , volume =

  225. [233]

    , journal =

    Yatracos, Yannis G. , journal =. Rates of convergence of minimum distance estimators and

  226. [234]

    and Tanner, Jared , journal =

    Donoho, David L. and Tanner, Jared , journal =. Counting the faces of randomly-projected hypercubes and orthants, with applications , volume =

  227. [235]

    and Tanner, Jared , journal =

    Donoho, David L. and Tanner, Jared , journal =. Counting faces of randomly projected polytopes when the projection radically lowers dimension , volume =

  228. [236]

    Proof of the optimality of the empirical star algorithm , year =

    Audibert, Jean-Yves , keywords =. Proof of the optimality of the empirical star algorithm , year =

  229. [237]

    and Birg

    Barron, Andrew R. and Birg. Risk bounds for model selection via penalization , volume =. Probability Theory and Related Fields , keywords =

  230. [238]

    Mendelson, Shahar , journal =. ``

  231. [239]

    Hypothesis testing by convex optimization , volume =

    Goldenshluger, Alexander and Juditsky, Anatoli and Nemirovski, Arkadi , journal =. Hypothesis testing by convex optimization , volume =

  232. [240]

    Entropy numbers, s-numbers, and eigenvalue problems , volume =

    Carl, Bernd , journal =. Entropy numbers, s-numbers, and eigenvalue problems , volume =

  233. [241]

    , journal =

    Milman, Vitali D. , journal =. Almost

  234. [242]

    Further inequalities for the (generalized)

    Alonso-Guti. Further inequalities for the (generalized). Communications in Contemporary Mathematics , keywords =

  235. [243]

    , institution =

    Donoho, David L. , institution =. Gelfand n -widths and the method of least squares , year =

  236. [244]

    , journal =

    Donoho, David L. , journal =. Statistical estimation and optimal recovery , volume =

  237. [245]

    Inequalities between intrinsic volumes , volume =

    McMullen, Peter , journal =. Inequalities between intrinsic volumes , volume =

  238. [246]

    , journal =

    Tsirel'son, Boris S. , journal =. A geometric approach to maximum likelihood estimation for infinite-dimensional location

  239. [247]

    , journal =

    Tsirel'son, Boris S. , journal =. A geometric approach to maximum likelihood estimation for infinite-dimensional

  240. [248]

    On the performance of kernel classes , volume =

    Mendelson, Shahar , journal =. On the performance of kernel classes , volume =

  241. [249]

    and Polyak, Boris T

    Nemirovski, Arkadii S. and Polyak, Boris T. and Tsybakov, Alexandre B. , journal =. Rate of convergence of nonparametric estimates of maximum-likelihood type , volume =

  242. [250]

    and Tsirel'son, Boris S

    Sudakov, Vladimir N. and Tsirel'son, Boris S. , journal =. Extremal properties of half-spaces for spherically invariant measures , volume =

  243. [251]

    , journal =

    Tsirel'son, Boris S. , journal =. A geometrical approach to maximum likelihood estimation for infinite-dimensional

  244. [252]

    Lattice points , year =

    Gritzmann, Peter and Wills, J. Lattice points , year =. Handbook of convex geometry. Volume B , keywords =

  245. [253]

    Uber affine Geometrie

    Blaschke, Wilhelm , journal =. Uber affine Geometrie

  246. [254]

    Un invariante afin para los cuerpos convexos del espacio de n dimensiones , volume =

    Santal. Un invariante afin para los cuerpos convexos del espacio de n dimensiones , volume =. Portugaliae Mathematica , keywords =

  247. [255]

    , journal =

    Bourgain, Jean and Milman, Vitaly D. , journal =. New volume ratio properties for convex symmetric bodies in

  248. [256]

    Optimal concentration of information content for log-concave densities , year =

    Fradelizi, Matthieu and Madiman, Mokshay and Wang, Liyao , booktitle =. Optimal concentration of information content for log-concave densities , year =

  249. [257]

    , journal =

    Sudakov, Vladimir N. , journal =. Geometric problems of the theory of infinite-dimensional probability distributions , volume =

  250. [258]

    Zur gitterpunktanzahl konvexer mengen , volume =

    Wills, J. Zur gitterpunktanzahl konvexer mengen , volume =. Elemente der Mathematik , keywords =

  251. [259]

    Hadwiger, Hugo , journal =. Das

  252. [260]

    Concentration of the information in data with log-concave distributions , volume =

    Bobkov, Sergey and Madiman, Mokshay , journal =. Concentration of the information in data with log-concave distributions , volume =

  253. [261]

    , journal =

    Vitale, Richard A. , journal =. The

  254. [262]

    and Rota, Gian-Carlo , keywords =

    Klain, Daniel A. and Rota, Gian-Carlo , keywords =. Introduction to Geometric Probability , year =

  255. [263]

    Stochastic and Integral Geometry , year =

    Schneider, Rolf and Weil, Wolfgang , keywords =. Stochastic and Integral Geometry , year =

  256. [264]

    , keywords =

    Gruber, Peter M. , keywords =. Convex and Discrete Geometry , year =

  257. [265]

    Aggregation via empirical risk minimization , volume =

    Lecu. Aggregation via empirical risk minimization , volume =. Probability Theory and Related Fields , keywords =

  258. [266]

    and Nourdin, Ivan and Peccati, Giovanni and Tropp, Joel A

    Lotz, Martin and McCoy, Michael B. and Nourdin, Ivan and Peccati, Giovanni and Tropp, Joel A. , booktitle =. Concentration of the intrinsic volumes of a convex body , volume =

  259. [267]

    Intrinsic volumes of polyhedral cones: a combinatorial perspective , volume =

    Amelunxen, Dennis and Lotz, Martin , journal =. Intrinsic volumes of polyhedral cones: a combinatorial perspective , volume =

  260. [268]

    Gordon, Yehoram , booktitle =. On

  261. [269]

    and Tropp, Joel A

    Amelunxen, Dennis and Lotz, Martin and McCoy, Michael B. and Tropp, Joel A. , journal =. Living on the edge: Phase transitions in convex programs with random data , volume =

  262. [270]

    and Lindenstrauss, Joram , journal =

    Johnson, William B. and Lindenstrauss, Joram , journal =. Extensions of

  263. [271]

    Cube slicing in

    Ball, Keith , journal =. Cube slicing in

  264. [272]

    The concept of duality in convex analysis, and the characterization of the

    Artstein-Avidan, Shiri and Milman, Vitali , journal =. The concept of duality in convex analysis, and the characterization of the

  265. [273]

    Gaussian model selection , volume =

    Birg. Gaussian model selection , volume =. Journal of the European Mathematical Society , keywords =

  266. [274]

    , journal =

    Haussler, David and Littlestone, Nick and Warmuth, Manfred K. , journal =. Predicting \ 0,1\ -functions on randomly drawn points , volume =

  267. [275]

    A function space view of bounded norm infinite width

    Ongie, Greg and Willett, Rebecca and Soudry, Daniel and Srebro, Nathan , journal =. A function space view of bounded norm infinite width

  268. [276]

    Learning bounded subsets of

    Mendelson, Shahar , journal =. Learning bounded subsets of

  269. [277]

    The law of large numbers and the central limit theorem in

    Hoffmann-J. The law of large numbers and the central limit theorem in. The Annals of Probability , keywords =

  270. [278]

    Finite-sample concentration of the multinomial in relative entropy , volume =

    Agrawal, Rohit , journal =. Finite-sample concentration of the multinomial in relative entropy , volume =

  271. [279]

    On the density of families of sets , volume =

    Sauer, Norbert , journal =. On the density of families of sets , volume =

  272. [280]

    and Nelson, Jelani , booktitle =

    Larsen, Kasper G. and Nelson, Jelani , booktitle =. Optimality of the

  273. [281]

    Fox, Daniel J. F. , journal =. A

  274. [282]

    Canonical barriers on convex cones , volume =

    Hildebrand, Roland , journal =. Canonical barriers on convex cones , volume =

  275. [283]

    The Cube -- A Window to Convex and Discrete Geometry , year =

    Zong, Chuanming , keywords =. The Cube -- A Window to Convex and Discrete Geometry , year =

  276. [284]

    The volume of convex bodies and

    Pisier, Gilles , keywords =. The volume of convex bodies and

  277. [285]

    , keywords =

    Artstein-Avidan, Shiri and Giannopoulos, Apostolos and Milman, Vitali D. , keywords =. Asymptotic Geometric Analysis, Part I , year =

  278. [286]

    Martingales in

    Pisier, Gilles , keywords =. Martingales in

  279. [287]

    and Zalgaller, Viktor A

    Burago, Yurii D. and Zalgaller, Viktor A. , keywords =. Geometric inequalities , year =

  280. [288]

    Convex bodies: The Brunn-Minkowski theory , year =

    Schneider, Rolf , edition =. Convex bodies: The Brunn-Minkowski theory , year =

  281. [289]

    and Boedihardjo, March T

    Bandeira, Afonso S. and Boedihardjo, March T. and van Handel, Ramon , journal =. Matrix concentration inequalities and free probability , volume =

  282. [290]

    Mutual information and minimum mean-square error in

    Guo, Dongning and Shamai, Shlomo and Verd. Mutual information and minimum mean-square error in. IEEE Transactions on Information Theory , keywords =

  283. [291]

    and Golitschek, Manfred v

    Lorentz, George G. and Golitschek, Manfred v. and Makovoz, Yuly , keywords =. Constructive Approximation: Advanced Problems , volume =

  284. [292]

    and Lorentz, George G

    DeVore, Ronald A. and Lorentz, George G. , keywords =. Constructive Approximation , volume =

  285. [293]

    n-Widths in Approximation Theory , year =

    Pinkus, Allan , keywords =. n-Widths in Approximation Theory , year =

  286. [294]

    Hadamard products and moments of random vectors , volume =

    Lata. Hadamard products and moments of random vectors , volume =. Advances in Mathematics , keywords =

  287. [295]

    and Nemirovsky, Arkadi S

    Nazin, Alexander V. and Nemirovsky, Arkadi S. and Tsybakov, Alexandre B. and Juditsky, Anatoli B. , journal =. Algorithms of robust stochastic optimization based on mirror descent method , volume =

  288. [296]

    Stochastic convex optimization , year =

    Shalev-Shwartz, Shai and Shamir, Ohad and Srebro, Nathan and Sridharan, Karthik , booktitle =. Stochastic convex optimization , year =

  289. [297]

    Robust estimation via robust gradient estimation , volume =

    Prasad, Adarsh and Suggala, Arun Sai and Balakrishnan, Sivaraman and Ravikumar, Pradeep , journal =. Robust estimation via robust gradient estimation , volume =

  290. [298]

    Conditions d'integrabilit

    Ledoux, Michel and Talagrand, Michel , journal =. Conditions d'integrabilit

  291. [299]

    Minsker, Stanislav , journal =

  292. [300]

    Isoperimetry and

    Ledoux, Michel , booktitle =. Isoperimetry and

Pith tools

Reviewed July 9, 2026 · model on record in the stance chip above.