betaFeedback

WorksComputer Science

Uniform convergence at optimal rate and scale

Abstract

Uniform convergence asks how many random samples suffice for empirical averages to approximate expectations simultaneously over an entire class of functions. It is a central concept in probability and a foundation of statistical learning. For binary-valued classes, the Vapnik–Chervonenkis theorem characterizes uniform convergence by finiteness of the VC dimension. For dimension dd, the sharp worst-case sample complexity for additive error ε\varepsilon is of order d/ε2d/\varepsilon^2. For real-valued functions, the analogous combinatorial parameter is the fat-shattering dimension, which depends on an approximation scale γγ. The challenge is to achieve error γ+εγ+\varepsilon with order d/ε2d/\varepsilon^2 samples, matching the sample bound for binary classes, where dd is now the fat-shattering dimension at scale γγ. Previous results achieved either the optimal scale or the optimal sample complexity, but not both. Our main result closes this gap, establishing the sharp sample complexity at the optimal scale. Along the way, we use several tools that may be of independent interest. The first concerns ε\varepsilon-nets with margin. A classical ε\varepsilon-net intersects every set of mass at least ε\varepsilon in a prescribed family; we allow the net to intersect a slightly enlarged version of each such set. Remarkably, this relaxation permits bounds independent of the ambient dimension in natural geometric settings. Second, we derive improved covering bounds in both discrete and real-valued settings: we extend Haussler's covering lemma to partial concept classes and use this extension to obtain ℓ2\ell_2 covers at the optimal scale for real-valued function classes. Third, we obtain a chaining bound at the optimal scale by aligning the covers so that successive approximations become increasingly close to one another.

Introduction

Suppose we wish to estimate the expectation of a bounded function ff under an unknown distribution. We can draw independent samples and average the values of ff on these samples. The Law of Large Numbers guarantees that this average converges to the expectation. Hoeffding’s Inequality [20] makes this quantitative: O(ε−2log⁡(1/δ))O(\varepsilon^{-2}\log(1/\delta)) samples suffice for error at most ε\varepsilon with confidence 1−δ1-\delta.

Suppose now we wish to estimate the expectations of MM such functions. A union bound gives the simultaneous guarantee with

O(log⁡M+log⁡(1/δ)ε2)O\left(\frac{\log M+\log(1/\delta)}{\varepsilon^2}\right)

samples. Thus, the number of functions enters only logarithmically.

Uniform convergence asks when a single sample can accurately estimate all expectations in an infinite class F\mathcal{F} of functions1. The required sample size may depend on F\mathcal{F}, the desired accuracy, and the confidence, but must work for every underlying distribution. Classes F\mathcal{F} for which the empirical averages uniformly approximate the expectations are called uniform Glivenko–Cantelli classes [15].

A classical example is estimating a cumulative distribution function (CDF). For independent samples S=(x1,…,xn)S=(x_1,\ldots,x_n) from a distribution DD on R\mathbb{R}, the true and empirical CDFs are

FD(t)=Px∼D[x≤t],F^S(t)=1n∑i=1n1{xi≤t}.F_D(t)=\mathbb{P}_{x\sim D}[x\le t],\qquad\widehat{F}_S(t)=\frac{1}{n}\sum_{i=1}^{n}\mathbf{1}_{\{x_i\le t\}}.

These two CDFs are the expectations and sample averages of the threshold functions ft(x)=1{x≤t}f_t(x)=\mathbf{1}_{\{x\le t\}}. Uniform convergence for this class means that the same sample estimates FD(t)F_D(t) accurately for all thresholds simultaneously. The Glivenko–Cantelli Theorem, proved in 1933 [18, 11], guarantees this accuracy with high probability for sufficiently large samples. An entire CDF can thus be estimated without a parametric model. The theorem is therefore sometimes called the Fundamental Theorem of Mathematical Statistics [RéN07, p.400].

Binary-valued functions. For classes of binary-valued functions, Vapnik and Chervonenkis [42] showed that uniform convergence holds precisely when the VC dimension is finite. This characterization also forms a major part of the Fundamental Theorem of PAC Learning [7]. Their proof also bounds the number of samples needed. For a class of VC dimension dd, their classical symmetrization argument, based on an independent ghost sample, gives

m(ε,δ)=O(dlog⁡(1/ε)+log⁡(1/δ)ε2).m(\varepsilon,\delta)=O\left(\frac{d\log(1/\varepsilon)+\log(1/\delta)}{\varepsilon^2}\right).

where ε\varepsilon is the approximation error and δ\delta is the allowed failure probability.

The extra logarithmic factor can be removed by combining chaining with bounds on the covering numbers of the function class. Dudley’s chaining method [16], Haussler’s covering estimates [19], and Talagrand’s bounds for empirical processes [41] yield the optimal worst-case sample complexity

Θ(d+log⁡(1/δ)ε2),d=VC⁡(F).\Theta\left(\frac{d+\log(1/\delta)}{\varepsilon^2}\right),\qquad d=\operatorname{VC}(\mathcal{F}).

This gives a tight theorem for binary classes, using techniques with applications in probability, functional analysis, and theoretical computer science. Our proof refines the covering and chaining tools, which we recall in § 2.

Real-valued functions. The binary case corresponds to estimating probabilities of events. More generally, we often want to estimate expectations of real-valued functions. For such classes, VC dimension is replaced by a scale-dependent extension, called the fat-shattering dimension [27].

Fix a scale γ>0\gamma>0. The dimension fat⁡γ(F)\operatorname{fat}_{\gamma}(\mathcal{F}) is the largest nn for which there are points x1,…,xnx_1,\ldots,x_n and thresholds r1,…,rnr_1,\ldots,r_n with the following property: for each of the 2n2^n ways to label the points x1,…,xnx_1,\ldots,x_n high or low, there is a function f∈Ff\in\mathcal{F} satisfying all the corresponding inequalities simultaneously:

f(xi)≥ri+γ2if xi is labeled high,f(x_i)\ge r_i+\frac{\gamma}{2}\quad\text{if }x_i\text{ is labeled high},
f(xi)≤ri−γ2if xi is labeled low.f(x_i)\le r_i-\frac{\gamma}{2}\quad\text{if }x_i\text{ is labeled low}.

The thresholds are chosen once, while ff may depend on the labeling.

For {0,1}\{0,1\}-valued classes and γ=1\gamma=1, this recovers VC dimension: realizing these patterns is exactly realizing every binary labeling. For real-valued classes, the fat-shattering dimension may decrease as the scale increases. Classical work showed that a bounded class satisfies uniform convergence exactly when its fat-shattering dimension is finite at every positive scale [1, 10]. These classical works reduce the study of uniform convergence to understanding what can be guaranteed at each scale:

If the fat-shattering dimension is finite at a scale γ\gamma, how accurately can we estimate the expectations of all functions in the class simultaneously?

Without information at finer scales, we cannot in general guarantee error below γ\gamma. The natural target is therefore γ+ε\gamma+\varepsilon, where the additional error ε\varepsilon can be reduced by drawing more samples. Our goal is to achieve this with the same sample complexity as in the binary case. We denote the required sample size, at confidence 1−δ1-\delta, by mγ(ε,δ)m_\gamma(\varepsilon,\delta). If this sample size is finite for every accuracy and confidence, the class satisfies γ\gamma-uniform convergence.

The classical bounds gave error Cγ+εC\gamma+\varepsilon, for a universal constant C>1C>1, with additional logarithmic factors in the sample complexity [1, 10]. Colomboni, Esposito, and Paudice [12], using bounds of Rudelson and Vershynin [37], obtained the optimal binary rate, but with error Cγ+εC\gamma+\varepsilon for a universal constant C>1C>1. The result of [AMM+26] reaches error γ+ε\gamma+\varepsilon, but retains additional logarithmic factors in the sample complexity. Thus, the optimal rate and the optimal scale were available separately.

Our contribution. We obtain both simultaneously:

Theorem 1.1: Main result Let F⊆[−1,+1]X\mathcal{F} \subseteq[-1,+1]^\mathcal{X}, let γ∈(0,1)\gamma\in(0,1), and suppose that d:=fat⁡γ(F)<∞d := \operatorname{fat}_{\gamma}(\mathcal{F}) < \infty. Then F\mathcal{F} satisfies γ\gamma-uniform convergence, with

mγ(ε,δ)=O(d+log⁡(1/δ)ε2),ε,δ∈(0,1).m_\gamma(\varepsilon,\delta) = O\left(\frac{d+\log(1/\delta)}{\varepsilon^2}\right), \qquad\varepsilon,\delta\in(0,1).

Moreover, for every γ∈(0,1)\gamma\in(0,1) and integer d≥1d \ge1, there is a class F\mathcal{F} with fat⁡γ(F)=d\operatorname{fat}_{\gamma}(\mathcal{F}) = d such that, for all sufficiently small ε,δ>0\varepsilon,\delta> 0,

mγ(ε,δ)=Ω(d+log⁡(1/δ)ε2).m_\gamma(\varepsilon,\delta) = \Omega\left(\frac{d+\log(1/\delta)}{\varepsilon^2}\right).

The theorem gives the optimal binary sample complexity at error γ+ε\gamma+\varepsilon. The matching lower bound shows that this cannot be improved in general using only the fat-shattering dimension at scale γ\gamma. Particular classes may admit stronger guarantees when more is known about their behavior at finer scales. Our proof combines covering bounds with chaining at optimal scale; the full proof appears in Section § 4.5.

In section § 2 we give an self-contained presentation overview of technical tools used to prove our main result. In section § 3 we give a proof overview of the main results in the paper. And finally, in section § 4 we give formal proofs for all results in the paper.

Technical contributions

Before turning to the proof of our uniform-convergence bound, we give a self-contained presentation of the technical tools behind it. Our aim is to explain the ideas in their natural settings and relate them to earlier work, so that the techniques can be understood beyond the present proof. This also brings out results that may be of independent interest, including some that are not used directly in our proof but illustrate the techniques more broadly.

We begin with ε\varepsilon-nets with margin and rounding fractional covers, then present covering bounds for partial binary classes and real-valued classes, and conclude with chaining at optimal scale in Euclidean space. The subsections can be read largely independently. In Section § 3, we return to uniform convergence and explain how these tools combine to prove our main result.

ε-nets with margin and rounding fractional covers

A basic question in combinatorics is how to choose a small set of points that intersects every sufficiently large set in a given family. Let X\mathcal{X} be a domain, let X⊂XX \subset\mathcal{X} be a set of nn points, and let U\mathcal{U} be a family of subsets of X\mathcal{X}. A set E⊆XE \subseteq\mathcal{X} is an ε\varepsilon-net for U\mathcal{U} with respect to XX if it intersects every U∈UU \in\mathcal{U} with ∣U∩X∣≥εn|U \cap X| \ge\varepsilon n. The Haussler–Welzl theorem [22] guarantees that

O(dεlog⁡1ε)O\left(\frac{d}{\varepsilon}\log\frac{1}{\varepsilon}\right)

points sampled independently and uniformly from XX form such a net with constant probability, where dd is the VC dimension of the set family U\mathcal{U}. Importantly, this bound is independent of nn.

Halfspaces with margin. Halfspaces in Rd\mathbb{R}^d provide a classical illustration of this theorem. Their VC dimension is d+1d+1, so the Haussler–Welzl theorem [22] guarantees that, given nn points, one can select O((d/ε)log⁡(1/ε))O((d/\varepsilon)\log(1/\varepsilon)) of them to meet every halfspace containing at least εn\varepsilon n points. Conversely, for d≥2d \ge2, there are point sets for which any such selection requires Ω(d/ε)\Omega(d/\varepsilon) points.2 The logarithmic gap between these upper and lower bounds has attracted considerable attention: Matoušek, Seidel, and Welzl removed it for d=2,3d=2,3 [MSW90], while Pach and Tardos showed that it is necessary already in dimension four [35]. Kupavskii, Mustafa, and Pach [26] subsequently established the matching lower bound Ω((d/ε)log⁡(1/ε))\Omega((d/\varepsilon)\log(1/\varepsilon)) for all d≥4d \ge4, settling the dependence on both parameters.3

We consider a natural relaxation of ε\varepsilon-nets that allows a margin: it suffices to hit each large halfspace after moving its boundary outward by a distance η\eta. Equivalently, place a ball of radius η\eta around each net point and require every large halfspace to intersect one of these balls. This margin viewpoint has been studied through partial concept classes [BCT25, IJN26]. It also appears in work on no-dimensional variants of Carathéodory’s, Helly’s, and Tverberg’s theorems and on weak ε\varepsilon-nets with margin for convex sets; see, e.g., [2, 25, 14, 21, 8, 24].

The following theorem gives a bound independent of the Euclidean dimension. It applies to ℓ2\ell_2 geometry in any real Hilbert space, including infinite-dimensional ones.

Theorem 2.1 (Halfspace nets with margin in Hilbert space). Let XX be a set of nn points in the unit ball4 of a real Hilbert space X\mathcal{X}, and let ε,η∈(0,1)\varepsilon,\eta\in(0,1). Then, there is a set E⊆XE \subseteq X of size O(1/(εη2))O(1/(\varepsilon\eta^2)) such that every closed halfspace containing at least εn\varepsilon n points of XX has distance at most η\eta from some point of EE.

The bound is optimal: for all sufficiently small ε,η>0\varepsilon,\eta>0, there is a finite set XX in the unit ball of ℓ2\ell_2 for which every such net has Ω(1/(εη2))\Omega(1/(\varepsilon\eta^2)) points.

This improves upon a result of Bourneuf, Charbit, and Thomassé [BCT25], who extended the Haussler–Welzl theorem to partial concept classes. Combining their extension with the classical bound of order 1/η21/\eta^2 on the VC dimension of halfspaces with margin η\eta gives a net of size O((εη2)−1log⁡(1/ε))O((\varepsilon\eta^2)^{-1}\log(1/\varepsilon)) in Hilbert space. Theorem 2.1 removes this logarithmic factor and gives the optimal dependence on ε\varepsilon and η\eta. Its full proof appears in Section 4.

It is interesting to compare this bound with the classical bounds without margin. Our bound is linear in 1/ε1/\varepsilon, whereas the classical bound has an extra factor of log⁡(1/ε)\log(1/\varepsilon) for d≥4d \ge4. Another difference is that our nets are weak, or external: their points may lie anywhere in the ambient space. They are not obtained by randomly sampling points from XX, as in the Haussler–Welzl construction. Indeed, independent uniform sampling from XX can still require the logarithmic factor even with a margin, as follows from lower bounds for margin-based generalization; see, e.g., [17 Lemma 11].

Beyond the Hilbert-space setting, Ivanov et al. also study margins for halfspaces in other normed spaces and for balls in LpL_p spaces [23]. It would be interesting to determine the optimal sizes of margin nets in these settings.

Rounding fractional covers

It is natural to study ε\varepsilon-nets with margin for other geometric families, such as polytopes of bounded complexity or even arbitrary convex sets. In each case, allowing a margin means slightly enlarging the set by moving its boundary outward. For balls, this simply increases the radius while keeping the center fixed. As with halfspaces, this relaxation yields dimension-independent bounds for balls. While the result holds in more general spaces, we present it in Hamming space. Both because this is the setting relevant to our proof and because it illustrates the applicability of margin nets in a more combinatorial setting. We use these nets to obtain a fractional extension of the familiar packing–covering relation.

For a finite set AA in a metric space (Z,ρ)(Z,\rho), let Pr(A)P_r(A) be the maximum number of pairwise disjoint closed balls of radius rr centered in AA, and let Cr(A)C_r(A) be the minimum number of radius-rr balls that cover AA, with centers anywhere in ZZ. These are the matching and transversal numbers of the ball family {Br(a):a∈A}\{B_r(a):a\in A\}: hitting every ball is equivalent to covering its center. The familiar packing–covering inequality is

C2r(A)≤Pr(A)≤Cr(A).C_{2r}(A) \le P_r(A) \le C_r(A).

Indeed, a maximal family of disjoint radius-rr balls yields a 2r2r-cover by the triangle inequality, while disjoint balls require distinct hitting points.

In a fractional cover, each center receives a nonnegative weight, and the centers within distance rr of each a∈Aa\in A must have total weight at least one. A fractional packing assigns nonnegative weights to the balls Br(a)B_r(a), with total weight at most one on the balls containing any given point. Linear-programming duality equates the minimum covering weight and maximum packing weight:

Cr∗(A)=Pr∗(A).C_r^*(A) = P_r^*(A).

The classical relation therefore gives C2r(A)≤Pr∗(A)=Cr∗(A)C_{2r}(A) \le P_r^*(A) = C_r^*(A).

We show that, in Hamming space, a fractional cover can be rounded with a radius increase of only 1+η1+\eta, at the expense of more centers. We use the normalized Hamming distance dHd_H on {0,1}n\{0,1\}^n, which is the fraction of coordinates on which two vectors differ, and write Br(x)={y:dH(x,y)≤r}B_r(x)=\{y:d_H(x,y)\le r\}. All packing and covering numbers below use the cube as the ambient space.

References

  1. [1]Noga Alon, Shai Ben-David, Nicolò Cesa-Bianchi, and David Haussler. Scale-sensitive dimensions, uniform convergence, and learnability. J. ACM, 44(4):615–631, July 1997.
  2. [2]Karim Adiprasito, Imre Bárány, Nabil H. Mustafa, and Tamás Terpai. Theorems of Carathéodory, Helly, and Tverberg without dimension. Discrete & Computational Geometry, 64(2):233–258, 2020.
  3. [3]Noga Alon, Steve Hanneke, Ron Holzman, and Shay Moran. A theory of pac learnability of partial concept classes. In 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS), pages 658–671. IEEE, 2022.arxiv.org/abs/2107.08444
  4. [6]Jean Bretagnolle, Didier Dacunha-Castelle, and Jean-Louis Krivine. Lois stables et espaces Lᵖ. Annales de l’I.H.P. Probabilités et statistiques, 2(3):231–259, 1966.DOI
  5. [7]Anselm Blumer, A. Ehrenfeucht, David Haussler, and Manfred K. Warmuth. Learnability and the vapnik-chervonenkis dimension. J. ACM, 36(4):929–965, October 1989.DOI
  6. [8]Polina Barabanshchikova, Grigory Ivanov, and Alexander Polyanskiĭ. Tight colorful no-dimensional Tverberg theorem. Combinatorica, 46(3), 2026. Article 25.arxiv.org/abs/2408.05814
  7. [9]Peter L. Bartlett, Sanjeev R. Kulkarni, and Steven R. Posner. Covering numbers for real-valued function classes. IEEE Transactions on Information Theory, 43(5):1721–1724, 1997.DOI
  8. [10]Peter L. Bartlett and Philip M. Long. Prediction, learning, uniform convergence, and scale-sensitive dimensions. Journal of Computer and System Sciences, 56(2):174–190, 1998.DOI
  9. [11]F. P. Cantelli. Sulla determinazione empirica delle leggi di probabilità. Giornale dell’Istituto Italiano degli Attuari, 4:421–424, 1933.
  10. [12]Roberto Colomboni, Emmanuel Esposito, and Andrea Paudice. An improved uni-form convergence bound with fat-shattering dimension. Information Processing Letters, 188:106539, 2025.
  11. [13]Bernard Chazelle. A note on Haussler’s packing lemma, 1992. Unpublished manuscript, Princeton University.
  12. [14]Aruni Choudhary and Wolfgang Mulzer. No-dimensional Tverberg theorems and algorithms. Discrete & Computational Geometry, 68(4):964–996, 2022.
  13. [15]Richard M Dudley, Evarist Giné, and Joel Zinn. Uniform and universal glivenko-cantelli classes. Journal of Theoretical Probability, 4(3):485–510, 1991.DOI
  14. [16]Richard M Dudley. Central limit theorems for empirical measures. The Annals of Probability, pages 899–929, 1978.DOI
  15. [17]Allan Grøn­lund, Lior Kamma, and Kasper Green Larsen. Near-tight margin-based generalization bounds for support vector machines. In Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pages 3779–3788. PMLR, 2020.
  16. [18]V. I. Glivenko. Sulla determinazione empirica delle leggi di probabilità. Giornale dell’Istituto Italiano degli Attuari, 4:92–99, 1933.
  17. [19]David Haussler. Sphere packing numbers for subsets of the boolean n-cube with bounded vapnik-chervonenkis dimension. Journal of Combinatorial Theory, Series A, 69(2):217–232, 1995.DOI
  18. [20]Wassily Hoeffding. Probability inequalities for sums of bounded random variables. Journal of the American Statistical Association, 58(301):13–30, 1963.
  19. [21]Sariel Har-Peled and Eliot W. Robson. No-dimensional Tverberg partitions revisited. In 19th Scandinavian Symposium and Workshops on Algorithm Theory (SWAT 2024), volume 294 of Leibniz International Proceedings in Informatics (LIPIcs), pages 26:1–26:14. Schloss Dagstuhl – Leibniz-Zentrum für Informatik, 2024.
  20. [22]David Haussler and Emo Welzl. Epsilon-nets and simplex range queries. In Proceedings of the second annual symposium on Computational geometry, pages 61–71, 1986.DOI
  21. [23]Grigory Ivanov, Attila Jung, and Márton Naszódi. The vc dimension of partial concept classes via radon’s theorem. arXiv preprint arXiv:2607.10751, 2026.arxiv.org/abs/2607.10751
  22. [24]Grigory Ivanov and Vladimir Kadets. Optimality of no-dimensional bounds in Banach spaces, 2026. arXiv:2607.04093.arxiv.org/abs/2607.04093
  23. [25]Grigory Ivanov. No-dimension Tverberg’s theorem and its corollaries in Banach spaces of type p. Bulletin of the London Mathematical Society, 53(2):631–641, 2021.DOI
  24. [26]Andrey Kupavskii, Nabil Mustafa, and János Pach. New lower bounds for ε-nets. In 32nd International Symposium on Computational Geometry (SoCG 2016), volume 51 of Leibniz International Proceedings in Informatics (LIPIcs), pages 54:1–54:16, 2016.
  25. [27]Michael J. Kearns and Robert E. Schapire. Efficient distribution-free learning of probabilistic concepts. Journal of Computer and System Sciences, 48(3):464–497, 1994.
  26. [28]Andrey Kupavskii and Nikita Zhivotovskiy. When are ε-nets small? Journal of Computer and System Sciences, 110:22–36, 2020.
  27. [29]Philip M. Long. On agnostic learning with {0, *, 1}-valued and real-valued hypotheses. In David P. Helmbold and Robert C. Williamson, editors, Computational Learning Theory, 14th Annual Conference on Computational Learning Theory, COLT 2001 and 5th European Conference on Computational Learning Theory, EuroCOLT 2001, Amsterdam, The Netherlands, July 16-19, 2001. Proceedings, volume 2111 of Lecture Notes in Computer Science, pages 289–302. Springer, 2001.DOI
  28. [30]Colin McDiarmid. On the method of bounded differences. In J. Siemons, editor, Surveys in Combinatorics, 1989: Invited Papers at the Twelfth British Combinatorial Conference, London Mathematical Society Lecture Note Series, pages 148–188. Cambridge University Press, Cambridge, 1989.DOI
  29. [32]Shahar Mendelson and Roman Vershynin. Entropy and the combinatorial dimension. Inventiones Mathematicae, 152(1):37–55, 2003.arxiv.org/abs/math/0203275
  30. [33]Chirag Pabbaraju. Optimal unambiguous DNFs and Alon–Saks–Seymour, 2026. arXiv:2608.02533.arxiv.org/abs/2608.02533
  31. [35]János Pach and Gábor Tardos. Tight lower bounds for the size of epsilon-nets. In Proceedings of the twenty-seventh annual symposium on Computational geometry, pages 458–463, 2011.
  32. [37]Mark Rudelson and Roman Vershynin. Combinatorics of random processes and sections of convex bodies. Annals of Mathematics, pages 603–648, 2006.arxiv.org/abs/math/0404192
  33. [38]Joel Spencer. Six standard deviations suffice. Transactions of the American mathematical society, 289(2):679–706, 1985.
  34. [39]Shai Shalev-Shwartz and Shai Ben-David. Understanding Machine Learning - From Theory to Algorithms. Cambridge University Press, 2014.
  35. [40]Nathan Srebro, Karthik Sridharan, and Ambuj Tewari. Smoothness, low noise and fast rates. In J. Lafferty, C. Williams, J. Shawe-Taylor, R. Zemel, and A. Culotta, editors, Advances in Neural Information Processing Systems, volume 23. Curran Associates, Inc., 2010.
  36. [41]Michel Talagrand. Sharper bounds for Gaussian and empirical processes. The Annals of Probability, 22(1):28–76, 1994.DOI
  37. [42]V. N. Vapnik and A. Ya. Chervonenkis. On the uniform convergence of relative frequencies of events to their probabilities. Theory of Probability & Its Applications, 16(2):264–280, 1971.

Paper details

vlnc:W_thk8772d
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…
Contents