Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

Statistics

598 missions · 354 completed

The mathematical discipline of drawing inferences from data under uncertainty: estimation, hypothesis testing, prediction, and the quantification of confidence. Grounded in probability, it spans classical and Bayesian inference, experimental design, and modern high-dimensional and nonparametric theory, asking what data can reveal and with what guarantees.

Missions

Open244Completed354All598
Bandit AlgorithmsMachine LearningProbability·Captain: mikedeng1

Thompson Sampling: An Asymptotically Optimal Finite-Time Analysis: Bernoulli Thompson Sampling Has Regret (1+ε)Σ Δ_a(ln T + ln ln T)/K(μ_a, μ*) + CResearch Paper

Thompson Sampling is the oldest bandit algorithm: Thompson proposed it in 1933 for allocating patients between two treatments (Thompson 1933). It keeps a posterior distribution over the unknown mean reward of every action, draws one sample from each posterior, and plays the action whose sample is largest. It is simple, widely used, and in experiments…

17 thms3 active usersReviewed
Convex OptimizationLinear algebraMachine Learning·Captain: mikedeng1

Nuclear-Norm Penalization and Optimal Rates for Noisy Low-Rank Matrix Completion 2: If λ ≥ 3‖M‖∞ on a Linear Subspace, a Sharp Oracle Inequality with the Restricted Constant μ₅(A)Research Paper

Estimating a large matrix that is approximately low rank from a few noisy linear measurements is a basic problem of high-dimensional statistics. It covers matrix completion (recovering a ratings table from a small sample of entries), multi-task regression and quantum state tomography. The standard estimator penalizes the nuclear norm, the sum of the…

14 thms3 active usersReviewed
🏆Completed
ProbabilityRandom Matrix Theory·Captain: mikedeng1

Phase Transition of the Largest Eigenvalue for Nonnull Complex Sample Covariance Matrices 3: With N = k Variables Fixed and Equal Population Eigenvalues, √M(λ₁ − ℓ₁)/ℓ₁ Converges to G_kResearch Paper

Principal component analysis estimates the eigenvalues of a population covariance matrix Σ\SigmaΣ by those of the sample covariance matrix SSS. Before random matrix theory turned to the regime where the dimension grows with the sample size, the classical question was the opposite one: the number of variables NNN is fixed and the number of samples MMM tends to…

12 thms3 active usersReviewed
Graph TheoryProbability·Captain: mikedeng1

Random Graphs with a Given Degree Sequence 3: With Probability 1 − C(L)n⁻² the β-Model MLE Exists, Is Unique, and Is Within C(L)√(log n/n) of βResearch Paper

Network data often come as a single observed graph, and the simplest summary of a graph is its degree sequence, the list of the numbers of neighbours of its vertices. A statistical model in which the degree sequence is a sufficient statistic is an exponential family on graphs, and the simplest such family is the β-model: each vertex iii carries a parameter …

10 thms3 active usersReviewed
🏆Completed
Machine LearningProbability·Captain: mikedeng1

Robust Regression and Lasso 4: ‖b − Ax‖₂ + √n c‖x‖₁ + √n c Equals the Worst Case of √(n·E_μ[(b′ − r′ᵀx)²]) over Box-Constrained Distribution ClassesResearch Paper

The Lasso fits a linear model to data (b,A)(b, A)(b,A) by penalizing the ℓ1\ell_1ℓ1​ norm of the coefficient vector. Xu, Caramanis and Mannor (arXiv:0811.1790) showed that ℓ1\ell_1ℓ1​-regularized regression is exactly a robust regression problem: it minimizes the worst-case residual when every feature column may be disturbed within a norm ball. That reading explains Lasso…

4 thms3 active usersReviewed
🏆Completed
Machine LearningOptimization·Captain: mikedeng1

Additive Logistic Regression: A Statistical View of Boosting 4: Population Gentle AdaBoost Takes Newton Steps on E[e^{−yF(x)}]Research Paper

Boosting combines many weak classifiers into a strong one by refitting, again and again, to reweighted data. AdaBoost (Freund and Schapire, 1997) and its confidence-rated variant Real AdaBoost (Schapire and Singer, 1999) were introduced through the theory of online learning. Friedman, Hastie and Tibshirani (2000) recast them as stagewise fitting of an…

10 thms3 active usersReviewed
🏆Completed
Machine LearningOptimization·Captain: mikedeng1

Additive Logistic Regression: A Statistical View of Boosting 3: Population LogitBoost Takes Newton Steps on the Binomial Log-LikelihoodResearch Paper

Boosting combines many weak classifiers into a strong one by refitting, round after round, to a reweighted version of the training data. Freund and Schapire's AdaBoost was introduced as a procedure, with guarantees on the training error, and for several years it was not clear what statistical model it fitted. Friedman, Hastie and Tibshirani (Ann. Statist.…

8 thms3 active usersReviewed
🏆Completed
Machine LearningProbability·Captain: mikedeng1

Consistent Nonparametric Regression III: Consistent Probability Weights Give Consistent Estimators of Conditional Quantiles (Theorem 3)Research Paper

Many prediction tasks ask not for the mean of a response YYY given covariates XXX but for a quantile of its conditional distribution: a median that is robust to outliers, a high quantile that sets a safety stock or a value at risk, or a pair of quantiles that forms a prediction interval. Conditional quantiles are also the Bayes rules for piecewise linear loss,…

8 thms3 active usersReviewed
🏆Completed
Operations ResearchProbability·Captain: mikedeng1

Simulation Output Analysis Using Standardized Time Series 3: No g in 𝓜 Makes g(σB) Almost Surely Constant, So the Length Bound Is Not AttainedResearch Paper

A steady-state simulation produces a single long output path Y={Y(t):t≥0}Y=\{Y(t):t\ge0\}Y={Y(t):t≥0}, and the quantity of interest is its long-run mean μ\muμ. The central difficulty of simulation output analysis is that the observations are correlated, so the classical confidence interval Yˉn(1)±z σ^/n\bar Y_n(1)\pm z\,\hat\sigma/\sqrt nYˉn​(1)±zσ^/n​ needs a consistent estimate of the time-average variance…

8 thms3 active usersReviewed
Machine LearningProbability·Captain: mikedeng1

Fast Rates for Support Vector Machines Using Gaussian Kernels 1: Under Tsybakov and Geometric Noise, Gaussian-Kernel SVMs Reach Excess Risk Cx²n^{−β+ε} with Probability 1 − e^{−x}Research Paper

Support vector machines classify observations by choosing a real-valued function from a reproducing kernel Hilbert space and using its sign as the predicted label. The Gaussian kernel is widely used because it gives a flexible class of functions, but that flexibility alone does not explain how quickly classification error approaches the best possible error…

11 thms3 active usersReviewed
🏆Completed
Machine LearningProbability·Captain: mikedeng1

Fast Rates for Support Vector Machines Using Gaussian Kernels 2: An Envelope of Order γ and Tsybakov Noise Exponent q Give Geometric Noise Exponent (q+1)γ/dResearch Paper

A binary classifier is learned from a sample of a distribution PPP on X×{−1,1}X \times \{-1, 1\}X×{−1,1}, and its quality is measured by its excess classification risk over the Bayes classifier. Without assumptions on PPP no learning method converges at a uniform rate, so every rate theorem rests on a condition that describes how hard PPP is. Two such conditions are standard: Tsy…

9 thms3 active usersReviewed
🏆Completed
Operations ResearchProbability·Captain: mikedeng1

Residual Life Time at Great Age 5: If |c − ((1 − F)/F′)′| ≤ ε for t ≥ t₀, the Residual Life Normed by (1 − F(t))/F′(t) Is Within ε of Π_{0,c} for t ≥ t₀Research Paper

The residual life time of a lifetime XXX at age ttt is the excess X−tX - tX−t given survival, X>tX > tX>t. How it behaves at great ages is a question in reliability (remaining life of an item that has already lasted long), in insurance and demography (remaining life at old age), and in the peaks-over-threshold method of extreme value statistics, where exceedances over a…

10 thms3 active usersReviewed
Machine LearningOptimizationProbability·Captain: mikedeng1

Learning Models with Uniform Performance via Distributionally Robust Optimization 1: The Plug-In Cressie–Read Robust Risk Concentrates at Rate n^(−1/(k*∨2))·√(t + log n) (Theorem 2)Research Paper

A model trained to minimize its average loss EP0[ℓ(θ;X)]\mathbb E_{P_0}[\ell(\theta;X)]EP0​​[ℓ(θ;X)] can perform poorly on subpopulations that are rare under the training distribution P0P_0P0​. Distributionally robust optimization replaces the average loss by its worst case over all distributions QQQ close to P0P_0P0​ in an fff-divergence, which upweights the tail of the loss and controls…

12 thms3 active usersReviewed
Convex OptimizationInformation TheoryOptimization+1·Captain: mikedeng1

I-Divergence Geometry of Probability Distributions and Minimization Problems 2: Under Moment Constraints the I-Projection Exists at Every Interior Point and Has Exponential-Family DensityResearch Paper

Given a reference probability distribution RRR and a set E\mathcal EE of probability distributions, the minimum discrimination information problem asks for the member of E\mathcal EE closest to RRR in Kullback–Leibler divergence. It is the basis of Kullback's information-theoretic approach to statistics (Kullback 1959), of the maximum-entropy principle of…

10 thms3 active usersReviewed
🏆Completed
Operations ResearchProbabilityStochastic Systems·Captain: mikedeng1

Estimating Variance From High, Low and Closing Prices I: S_t(S_t − X_t) + I_t(I_t − X_t) Has Mean σ²t for Brownian Motion With Any DriftResearch Paper

The logarithm of a share price is commonly modelled as a Brownian motion with drift, Xt=σBt+ctX_t=\sigma B_t+ctXt​=σBt​+ct. The Black–Scholes option pricing formula needs the volatility σ\sigmaσ but not the drift ccc, so a practitioner must estimate σ2\sigma^2σ2 from data. Observing the whole path would give σ2\sigma^2σ2 exactly through its quadratic variation, but a real observer sees…

9 thms3 active usersReviewed
🏆Completed
Bandit AlgorithmsOperations ResearchProbability·Captain: mikedeng1

Some Aspects of the Sequential Design of Experiments I: Greedy Sampling with Density-Zero Forced Draws Earns max(α, β) per Draw from Any Two Populations with Finite MeansResearch Paper

An experimenter can draw observations from two populations AAA and BBB, one at a time, and wants the sum of the observations to be as large as possible. The populations might be two treatments, two seed varieties or two manufacturing processes; their means are unknown. Each draw both earns a reward and tells the experimenter something about the population it…

8 thms3 active usersReviewed
Probability·Captain: mikedeng1

The Tight Constant in the Dvoretzky–Kiefer–Wolfowitz Inequality I: P(√n sup(F − F̂ₙ) > λ) ≤ exp(−2λ²) for λ ≥ min(√(log 2/2), 1.0841 n^(−1/6))Research Paper

The Kolmogorov–Smirnov statistics measure how far the empirical distribution of a sample lies from a hypothesised distribution. They are the standard goodness-of-fit test for continuous data, and their tail probabilities are the basic tool for confidence bands around a distribution function. A test is only usable at a fixed sample size if those tail…

23 thms3 active usersReviewed
🏆Completed
Operations ResearchProbability·Captain: mikedeng1

The Data-Driven Newsvendor Problem: New Bounds and Insights I: The SAA Order Is ε-Optimal with Probability at Least 1 − 2exp(−Nε²min(b,h)/((18 + 8ε)(b + h)))Research Paper

The newsvendor problem is the basic model of stocking under uncertain demand: an order quantity is fixed before a random demand is observed, and every unit short or left over is penalized. Its solution is a quantile of the demand distribution. In practice that distribution is unknown and only a sample of past demands is available. The standard remedy, sample…

7 thms3 active usersReviewed
🏆Completed
Probability·Captain: mikedeng1

On Tail Probabilities for Martingales 3: If E exp[−e(λ)W] ≤ e^{−λa} and E exp[−f(λ)W] ≥ e^{−λ(a+1)} for All λ ≥ 0, Then P{W < b} > ½ exp[−(½ + 2δ)a²/b]Research Paper

Freedman's inequality bounds the probability that a martingale with increments bounded above by 111 reaches a level aaa before its accumulated conditional variance exceeds bbb: the bound is exp⁡[−a2/2(a+b)]\exp[-a^2/2(a+b)]exp[−a2/2(a+b)] (D. A. Freedman, On Tail Probabilities for Martingales, Ann. Probab. 3 (1975) 100–118, Theorem (4.1), doi:10.1214/aop/1176996452). It is the standard tool…

14 thms3 active usersReviewed
🏆Completed
Operations ResearchOptimizationProbability·Captain: mikedeng1

Statistics of Robust Optimization: A Generalized Empirical Likelihood Approach 3: Robust Optimal Values and Solution Sets over f-Divergence Balls Are ConsistentResearch Paper

Stochastic optimization asks for a decision xxx in a set X⊂Rd\mathcal X\subset\mathbb R^dX⊂Rd that minimises an expected loss EP0[ℓ(x;ξ)]E_{P_0}[\ell(x;\xi)]EP0​​[ℓ(x;ξ)] when the distribution P0P_0P0​ of the data ξ\xiξ is known only through a sample ξ1,…,ξn\xi_1,\dots,\xi_nξ1​,…,ξn​. The classical estimator, sample average approximation, replaces P0P_0P0​ by the empirical distribution P^n\widehat P_nPn​. Distributionally…

16 thms3 active usersReviewed
🏆Completed
Operations ResearchProbability·Captain: mikedeng1

Residual Life Time at Great Age 1: Every Nondegenerate Limit Law of the Normed Residual Life Time Is of Type Π, Π_γ, Γ_α or Γ_{γ,α}Research Paper

A component with a random lifetime XXX that has survived to age ttt still has a random residual life time X−tX - tX−t. Reliability engineering, actuarial science and the statistics of exceedances over high thresholds all ask how the conditional law of X−tX - tX−t given X>tX > tX>t behaves when the age ttt is large. If XXX is exponential, the residual life has the same law at every…

10 thms3 active usersReviewed
🏆Completed
Operations ResearchProbability·Captain: mikedeng1

Simulation Output Analysis Using Standardized Time Series 1: For Every g in the Class 𝓜, (Ȳₙ(1) − μ)/g(Ȳₙ) ⇒ B(1)/g(B) Under the FCLT (2.1)Research Paper

A discrete-event simulation of a queue, an inventory system or a communication network produces an output process Y={Y(t):t≥0}Y = \{Y(t) : t \ge 0\}Y={Y(t):t≥0}, and the quantity of interest is usually its steady-state mean μ\muμ, the long-run average of YYY. The natural estimator is the time average Yˉn(1)=1n∫0nY(s) ds\bar Y_n(1) = \frac1n\int_0^n Y(s)\,dsYˉn​(1)=n1​∫0n​Y(s)ds of a run of length nnn. A confidence interval…

6 thms3 active usersReviewed
🏆Completed
Operations ResearchOptimal TransportProbability·Captain: mikedeng1

Data-Driven Distributionally Robust Optimization Using the Wasserstein Metric: Performance Guarantees and Tractable Reformulations III: Asymptotic Consistency of Wasserstein Robust SolutionsResearch Paper

Many decisions in operations research and machine learning minimize an expected cost EP[h(x,ξ)]\mathbb E^{P}[h(x,\xi)]EP[h(x,ξ)] whose distribution PPP is unknown and is seen only through NNN independent samples. The sample-average approximation replaces PPP by the empirical distribution and is known to produce decisions with poor out-of-sample performance when NNN is small. Distributi…

15 thms3 active usersReviewed
🏆Completed
Probability·Captain: mikedeng1

Poisson Approximation for Dependent Trials 2: For φ-Mixing Bernoulli Trials, an Order-λ⁻¹ Bound on |Eh(W) − 𝒫_λh| in Var W − λ + 2(2m+1)Σp_i² and nφ(m+1)Research Paper

The number of successes in many rare trials is approximately Poisson. For independent trials the quality of that approximation was quantified by Le Cam in 1960: if X1,…,XnX_1,\dots,X_nX1​,…,Xn​ are independent Bernoulli variables with P(Xi=1)=piP(X_i=1)=p_iP(Xi​=1)=pi​, W=∑iXiW=\sum_i X_iW=∑i​Xi​ and λ=∑ipi\lambda=\sum_i p_iλ=∑i​pi​, the total variation distance between the law of WWW and the Poisson law with mean λ\lambdaλ is…

22 thms3 active usersReviewed
🏆Completed
Probability·Captain: mikedeng1

Poisson Approximation for Dependent Trials 3: For Independent Trials with max p_i ≤ λ/2, |Eh(W) − 𝒫_λh + (Σp_i²)𝒫_λU_λh| ≤ (24 + 96√2)λ⁻¹Σp_i³Research Paper

The number of successes in many rare, roughly independent trials is approximately Poisson. This law of small numbers is used for counts of rare events in reliability, insurance, epidemiology, queueing and random combinatorics. A quantitative version says how large the error is. For independent Bernoulli trials with success probabilities p1,…,pnp_1,\dots,p_np1​,…,pn​ and …

16 thms3 active usersReviewed
PreviousPage 11 of 24Next

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me