Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in

Get started

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

Machine Learning

886 missions · 536 completed

The science of systems that learn from data and experience. Its scope runs from the statistical and mathematical foundations of learning, including generalization, expressivity, and computational limits, through the design of learning algorithms, deep learning, reinforcement learning, and probabilistic methods, to the empirical study of large models and the trustworthiness, interpretability, and societal impact of learned systems.

Missions

Open350Completed536All886
🏆Completed
ProbabilityStatistics·Captain: mikedeng1

Rademacher and Gaussian Complexities: Risk Bounds and Structural Results 5: Kernel Expansions with α′Kα ≤ B² Have Rademacher and Gaussian Complexity at Most 2B√(E k(X,X)/n)Research Paper

Kernel methods, such as support vector machines, predict with functions of the form x↦∑iαik(x,xi)x \mapsto \sum_i \alpha_i k(x, x_i)x↦∑i​αi​k(x,xi​): finite expansions of a fixed similarity function kkk centred at data points. Their statistical behaviour is governed by the size of the class of such expansions that the method searches. Bartlett and Mendelson, in Rademacher and Gaussian…

8 thms3 active usersReviewed
🏆Completed
OptimizationReinforcement Learning·Captain: mikedeng1

On the Theory of Policy Gradient Methods: Optimality, Approximation, and Distribution Shift 1: Projected Gradient Ascent on the Simplex Is ε-Optimal After 64γ|S||A|D∞²/((1−γ)⁶ε²) IterationsResearch Paper

Policy gradient methods optimize a parameterized policy of a Markov decision process by gradient ascent on its expected discounted return. They are among the most widely used methods in reinforcement learning, from REINFORCE (Williams 1992) and the policy gradient theorem to natural policy gradient and trust-region methods. The objective is not concave in…

18 thms3 active usersReviewed
🏆Completed
Optimal TransportOptimizationStatistics·Captain: mikedeng1

Robust Wasserstein Profile Inference and Applications to Machine Learning 1: Square-Root LASSO Is Wasserstein DRO — the Worst-Case Squared Loss over D_c(P, P_n) ≤ δ Equals (√MSE_n(β) + √δ‖β‖_p)²Research Paper

Regularized least squares is the standard tool of high-dimensional linear regression. The square-root LASSO of Belloni, Chernozhukov and Wang (Biometrika, 2011) minimizes MSEn(β)+λ∥β∥1\sqrt{\mathrm{MSE}_n(\beta)} + \lambda\|\beta\|_1MSEn​(β)​+λ∥β∥1​. Unlike the LASSO, its optimal regularization parameter does not depend on the unknown noise level. Regularization is usually justified…

14 thms3 active usersReviewed
🏆Completed
Convex OptimizationOperations ResearchOptimization·Captain: mikedeng1

Oracle-Based Robust Optimization via Online Learning 1: The Dual-Subgradient Meta-Algorithm Returns a 2ε-Approximate Robust Solution or Certifies Infeasibility within ⌈G²D²/ε²⌉ Oracle CallsResearch Paper

Robust optimization protects a decision against every realization of uncertain data in a prescribed uncertainty set. The standard approach replaces the uncertain constraints by a deterministic robust counterpart and solves that counterpart directly (Ben-Tal, El Ghaoui, Nemirovski, Robust Optimization, 2009). The counterpart is often a harder problem than…

8 thms3 active usersReviewed
🏆Completed
ProbabilityStatistics·Captain: mikedeng1

Rademacher and Gaussian Complexities: Risk Bounds and Structural Results 4: A Fixed Boolean Combination of k Classes Has Gaussian Complexity at Most 2 Σ_j G_n(F_j)Research Paper

Data-dependent risk bounds in statistical learning replace combinatorial quantities such as the VC dimension with averages of how well a function class can fit random noise on the observed sample. Bartlett and Mendelson's article (JMLR 3, 2002) established these averages, the Rademacher and Gaussian complexities, as a general tool: a risk bound (their…

8 thms3 active usersReviewed
🏆Completed
Operations ResearchOptimizationReinforcement Learning·Captain: mikedeng1

Twice Regularized MDPs and the Equivalence Between Robustness and Regularization 2: The Greedy Policy of the R2 Optimal Value Is the Unique Optimal R2 PolicyResearch Paper

A robust Markov decision process (robust MDP) evaluates a policy against the worst transition kernel and reward in an uncertainty set around a nominal model (P0,r0)(P_0, r_0)(P0​,r0​). It is the standard model for planning when the dynamics are estimated from data (Iyengar 2005; Nilim and El Ghaoui 2005; Wiesemann, Kuhn and Rustem 2013). Its Bellman update contains an…

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

The Best of Both Worlds: Stochastic and Adversarial Bandits: SAO Has Pseudo-Regret O(K log K log²β/Δ) on Stochastic Rewards and Regret Õ(√(nK)) Against Adaptive AdversariesResearch Paper

In a multi-armed bandit problem a learner chooses one of KKK actions in each of nnn rounds and observes only the reward of the chosen action. Two models of the rewards have separate theories. In the stochastic model, each arm pays independent draws from a fixed distribution; algorithms such as UCB1 (Auer, Cesa-Bianchi & Fischer 2002) have regret of order …

20 thms3 active usersReviewed
🏆Completed
Bandit AlgorithmsOperations Research·Captain: mikedeng1

Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems I: Pseudo-Regret of (α, ψ)-UCBTextbook

The stochastic multi-armed bandit is the basic model of sequential decisions under uncertainty with partial feedback: a forecaster repeatedly picks one of KKK options and observes only the reward of the option it picked. It models clinical trials, ad placement, routing and dynamic pricing, and is the building block of many reinforcement-learning algorithms.…

12 thms3 active usersReviewed
🏆Completed
ProbabilityStatistics·Captain: mikedeng1

The Sample Complexity of Pattern Classification with Neural Networks: The Size of the Weights is More Important than the Size of the Network III: Sigmoid Networks with Small Weights GeneralizeResearch Paper

A classifier built from a neural network produces a real score and predicts a binary label from its sign. A network can have many hidden units, so a guarantee based only on the number of parameters can be uninformative even when its output weights are small. Bartlett's 1998 paper asks whether a classifier's margin on training examples and the total…

20 thms3 active usersReviewed
🏆Completed
Convex OptimizationOptimizationProbability+1·Captain: mikedeng1

Variance-based Regularization with Convex Objectives IV: Fast Rates for Approximate Robust Minimizers under a Growth ConditionResearch Paper

In stochastic optimization and statistical learning one chooses a parameter θ\thetaθ from a set Θ⊆Rd\Theta\subseteq\mathbb R^dΘ⊆Rd to make the risk R(θ)=EP[ℓ(θ;X)]R(\theta)=\mathbb E_P[\ell(\theta;X)]R(θ)=EP​[ℓ(θ;X)] small, having seen only a sample X1,…,XnX_1,\dots,X_nX1​,…,Xn​ from PPP. Generalization bounds suggest trading empirical risk against its standard deviation, but the variance-penalized objective is…

12 thms3 active usersReviewed
🏆Completed
OptimizationProbabilityStatistics·Captain: mikedeng1

Variance-based Regularization with Convex Objectives II: A Covering-Number Certificate and Oracle Inequality for the Robust MinimizerResearch Paper

Empirical risk minimization (ERM) chooses, from a class F\mathcal FF of loss functions, the one with the smallest average loss on a sample X1,…,XnX_1,\dots,X_nX1​,…,Xn​. Its standard guarantees bound the excess population risk by a term of order 1/n1/\sqrt n1/n​, whatever the variance of the losses. When good functions in F\mathcal FF have small variance, a better trade-off is…

11 thms3 active usersReviewed
🏆Completed
Linear OptimizationProbabilityStatistics·Captain: mikedeng1

The Dantzig Selector: Statistical Estimation When p Is Much Larger than n 2: Oracle Inequality within a Logarithmic Factor of the Ideal Mean Squared ErrorResearch Paper

Many regression problems have far more unknown coefficients ppp than observations nnn: gene expression studies with tens of samples and thousands of genes, imaging from few measurements, nonparametric curve recovery from a finite number of noisy samples. Estimation is hopeless in general, but becomes possible when the parameter is sparse, that is, has few…

11 thms3 active usersReviewed
🏆Completed
AnalysisFunctional Analysis·Captain: mikedeng1

Theory of Reproducing Kernels II: One Reproducing Kernel Class Contains Another iff a Multiple of Its Kernel Dominates the Other KernelResearch Paper

A reproducing kernel Hilbert space (RKHS) is a Hilbert space of functions in which evaluation at each point is a continuous functional. Each such space is determined by its kernel, and in practice spaces are usually given by their kernels: Gaussian or Sobolev kernels in kernel methods and Gaussian-process regression, the Bergman and Szegő kernels in complex…

14 thms3 active usersReviewed
🏆Completed
Convex OptimizationProbabilityStatistics·Captain: mikedeng1

Stability and Generalization 4: Relative-Entropy Regularization of Mixtures Has Uniform Stability M²/(λm)Research Paper

A learning algorithm generalizes when its error on fresh data is close to its error on the training sample. Bousquet and Elisseeff (JMLR 2, 2002) showed that a single property of the algorithm, uniform stability, controls this gap with exponential concentration: if removing any one example from a training set of size mmm changes the loss of the output at…

18 thms3 active users
🏆Completed
ProbabilityStatistics·Captain: mikedeng1

Stability and Generalization 2: Exponential Generalization Bounds for Uniformly Stable AlgorithmsResearch Paper

A learning algorithm is judged by its generalization error: the expected loss of the hypothesis it outputs on a fresh example. That quantity depends on an unknown distribution, so it is estimated from the training data, by the empirical error (the average loss on the training set) or the leave-one-out error (the average loss on each training point of the…

13 thms3 active usersReviewed
🏆Completed
Numerical AnalysisOptimization·Captain: mikedeng1

Gradient Convergence in Gradient Methods with Errors I: With Deterministic Errors Proportional to the Stepsize, Either f(x_t) → −∞ or f(x_t) Converges and ∇f(x_t) → 0Research Paper

Gradient methods are the workhorse of large-scale nonlinear optimization and of the training of statistical models. In practice the direction actually used is rarely the exact negative gradient: it may be scaled, computed incrementally one data component at a time, or perturbed by approximation error. The classical convergence theory of such methods often…

6 thms3 active usersReviewed
🏆Completed
Dynamic ProgrammingMarkov ChainReinforcement Learning·Captain: mikedeng1

Reinforcement Learning: An Introduction IX: Average Reward and the Futility of Discounting in Continuing ProblemsTextbook

Reinforcement learning formulates control as maximizing reward accumulated over time. For continuing tasks, where interaction never terminates, the standard textbook objective is the discounted return with a discount rate γ<1\gamma < 1γ<1. Chapter 10 of Sutton and Barto, Reinforcement Learning: An Introduction (2nd ed., MIT Press, 2018), argues that once values…

11 thms3 active usersReviewed
🏆Completed
Convex OptimizationOptimization·Captain: mikedeng1

SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives II: The 4n/k Rate of the Averaged Iterate without Strong ConvexityResearch Paper

Many problems in statistics and machine learning minimize an average of nnn losses, one per data point, plus a regularizer: least squares, logistic regression, and their ℓ1\ell_1ℓ1​- or ℓ2\ell_2ℓ2​-penalized versions. When nnn is large, a full gradient costs nnn component gradients, while stochastic gradient descent, which uses one component per step, needs decreasing step…

11 thms3 active usersReviewed
🏆Completed
Convex OptimizationOptimization·Captain: mikedeng1

SAGA: A Fast Incremental Gradient Method With Support for Non-Strongly Convex Composite Objectives I: Linear Convergence under Strong ConvexityResearch Paper

Many problems in machine learning and statistics are finite sums: an empirical risk f(x)=1n∑i=1nfi(x)f(x)=\frac1n\sum_{i=1}^n f_i(x)f(x)=n1​∑i=1n​fi​(x) over nnn data points, often plus a regulariser hhh such as an ℓ1\ell_1ℓ1​ penalty. When nnn is large, a full gradient of fff costs nnn component gradients, while stochastic gradient descent uses one component per step but converges only sublinearly because…

11 thms3 active usersReviewed
🏆Completed
Dynamic ProgrammingReinforcement Learning·Captain: mikedeng1

Reinforcement Learning: An Introduction IV: Policy Iteration for ε-Soft PoliciesTextbook

Policy iteration alternates two steps: evaluate the current policy, then replace it by a policy that is greedy with respect to the evaluated action values. The policy improvement theorem guarantees that each greedy step does not make the policy worse, and that the process stops only at an optimal policy. When the action values are estimated from experience…

11 thms3 active usersReviewed
🏆Completed
Linear OptimizationProbabilityStatistics·Captain: mikedeng1

The Dantzig Selector: Statistical Estimation When p Is Much Larger than n 1: ℓ2 Error Bound for Sparse Parameters under the Uniform Uncertainty PrincipleResearch Paper

In many statistical applications the number of unknown parameters ppp is far larger than the number of observations nnn: gene-expression studies with tens of samples and thousands of genes, imaging problems with fewer measurements than pixels, and nonparametric curve estimation from finitely many noisy samples. Least squares is useless in this regime, since the…

9 thms3 active usersReviewed
🏆Completed
Markov ChainOptimizationReinforcement Learning·Captain: mikedeng1

Reinforcement Learning: An Introduction XII: The Policy Gradient TheoremTextbook

Policy gradient methods learn a parameterized policy π(a∣s,θ)\pi(a \mid s, \theta)π(a∣s,θ) directly, by stochastic gradient ascent on a scalar performance measure J(θ)J(\theta)J(θ), instead of deriving the policy from learned action values. They are how reinforcement learning handles continuous action spaces, stochastic optimal policies and prior knowledge built into the policy's…

14 thms3 active usersReviewed
🏆Completed
OptimizationProbability·Captain: mikedeng1

Gradient Convergence in Gradient Methods with Errors II: With Zero-Mean Stochastic Errors, Almost Surely Either f(x_t) → −∞ or f(x_t) Converges and ∇f(x_t) → 0Research Paper

Stochastic gradient methods minimize a function fff when only noisy estimates of its gradient are available: each step moves along a descent direction corrupted by random noise. They are the standard training algorithm for neural networks and the basic tool of stochastic approximation, and the question every user faces is what can be guaranteed when fff is nonco…

4 thms3 active usersReviewed
🏆Completed
Markov ChainReinforcement Learning·Captain: mikedeng1

Reinforcement Learning: An Introduction VII: The Error Reduction Property of n-step ReturnsTextbook

Temporal-difference (TD) learning estimates the value of a policy by moving a current estimate toward a target built from observed rewards and from the estimate itself. Chapter 7 of Sutton and Barto's Reinforcement Learning: An Introduction (2nd ed., MIT Press, 2018) interpolates between the two extreme targets of the preceding chapters: the one-step TD…

10 thms3 active usersReviewed
🏆Completed
Algorithmic Game TheoryMechanism Design·Captain: mikedeng1

How Much Data Is Sufficient to Learn High-Performing Algorithms? Generalization Guarantees for Data-Driven Algorithm Design 2: Neutral Affine Maximizers Have Pseudo-Dimension at Least ⌊n/2⌋Research Paper

In data-driven algorithm design, an algorithm or mechanism comes with a vector of tunable parameters ρ\rhoρ, and the parameters are chosen by optimizing average performance over sample instances drawn from an unknown distribution. How many samples suffice for the empirical optimum to be near-optimal in expectation is governed by the pseudo-dimension of the…

6 thms3 active usersReviewed
PreviousPage 19 of 36Next

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