Paper improves risk bounds for nonconvex-strongly-concave minimax problems.
problem Achieving sharper risk bounds for nonconvex-strongly-concave minimax problems.
method Using uniform localized convergence to derive high probability generalization error bounds.
result Derives n times faster excess primal risk bounds for popular algorithms.
The paper bounds the mean absolute error in DNN vector-to-vector regression.
problem Bounding the mean absolute error in deep neural network based vector-to-vector regression.
method Error decomposition techniques in statistical learning theory and non-convex optimization theory were used to derive upper bounds for approximation, estimation, and optimization errors.
result Theoretical upper bounds for mean absolute error in DNN vector-to-vector regression were derived and validated experimentally.
Full-batch GD achieves generalization close to any stationary point with fewer assumptions.
problem Generalization and excess risk bounds for smooth losses, including non-Lipschitz and nonconvex cases.
method Path-dependent analysis of GD's generalization error, focusing on optimization error and stability.
result Generalization error is tightly bound in terms of optimization error and iteration count, bypassing common assumptions.
Sharp bounds found on expert error in binary advice aggregation.
problem Aggregating binary advice from conditionally independent experts.
method Sharp upper and lower bounds on optimal error probability in asymmetric case.
result Sharp bounds recover and sharpen known results in symmetric case.
Error bounds, which refer to inequalities that bound the distance of vectors in a test set to a given set by a residual function, have proven to be extremely useful in analyzing the convergence rates of a host of iterative methods for solving optimization problems. In this paper, we present a new framework for establis…
Adversarial training achieves optimal test error for shallow networks.
problem Achieving optimal adversarial test error for general data distributions.
method Applying new Rademacher complexity bounds and properties of optimal adversarial predictors.
result Adversarial training can achieve optimal adversarial test error for general data distributions.
A new error bound improves safety in Bayesian optimization.
problem Ensuring safety in Bayesian optimization with probabilistic models.
method Introducing a novel error bound using Wiener kernel regression for Gaussian processes and noise.
result The new error bound provides larger safety regions than previous methods.
Study error bounds and optimal schedules for Masked Diffusions with factorized approximations.
problem Analyzing trade-offs between computation and accuracy in Masked Diffusion Models.
method Provided general error bounds and identified optimal schedules based on data distribution information profiles.
result Identified optimal schedule sizes for Masked Diffusion Models.
Novel bounds for SGLD show generalization error decreases with more samples.
problem Understanding the generalization error of SGLD in non-convex optimization.
method Information-theoretic approach focusing on Kullback-Leibler divergence and sub-exponential loss function.
result Time-independent generalization bounds for SGLD, independent of step size and number of iterations.
This research evaluates generalization measures in deep learning.
problem Understanding why deep learning models generalize well despite small training error.
method Empirical evaluation of generalization bounds and measures.
result Generalization measures should be evaluated using distributional robustness.
Study problem-dependent rates in statistical learning theory, achieving optimal generalization error bounds.
problem Generalization error in statistical learning theory.
method Uniform localized convergence framework.
result Optimal generalization error bounds for various learning problems.
Many machine learning tasks can be formulated as Regularized Empirical Risk Minimization (R-ERM), and solved by optimization algorithms such as gradient descent (GD), stochastic gradient descent (SGD), and stochastic variance reduction (SVRG). Conventional analysis on these optimization algorithms focuses on their conv…
Improved BO algorithms reduce prediction error under Gaussian noise.
problem Reducing prediction error in Bayesian optimization with Gaussian noise.
method Established new prediction error bounds for Gaussian process under frequentist setting.
result Proved improved convergence rates of cumulative regret for GP-UCB and GP-TS.
Develops bounds predicting deep learning generalization using optimal transport.
problem Discrepancy between theoretical error bounds and empirical observations in deep learning.
method Margin-based generalization bounds with optimal transport costs.
result Theoretical bounds robustly predict generalization error on large datasets.
Sharp 2-Wasserstein bounds for DDPMs derived from Föllmer process.
problem Sampling error bounds for DDPMs in 2-Wasserstein distance.
method Lipschitz-type conditions on score function, Föllmer process, and log-concave target distributions.
result Sharp upper bounds for DDPMs in 2-Wasserstein distance, optimal in dimension and steps.
New bounds for SGLD show error decreases with more data.
problem Establishing generalization error bounds for SGLD in non-convex settings.
method Using dissipativity, smoothness, and uniform stability, time-independent bounds are derived.
result Error bounds decay to zero as sample size increases.
New algorithm improves regression error bounds and accelerates performance for low noise.
problem Nonparametric least square regression in RKHS with optimal error bounds.
method Kernel Truncated Randomized Ridge Regression (KTRRR) with optimal generalization error bounds.
result Faster finite-time and asymptotic rates on low noise problems.
New research shows many batch selection methods for training work just as well as full batch training.
problem Finding optimal batch selection methods for training.
method Analysis of mini-batch Gradient Descent (GD) and Stochastic GD (SGD) with various batch selection rules.
result All mini-batch schedules, including deterministic ones, generalize optimally for smooth Lipschitz-convex/nonconvex/strongly-convex loss functions.
In this paper we study the stability and its trade-off with optimization error for stochastic gradient descent (SGD) algorithms in the pairwise learning setting. Pairwise learning refers to a learning task which involves a loss function depending on pairs of instances among which notable examples are bipartite ranking,…
Algorithmic stability is a classical approach to understanding and analysis of the generalization error of learning algorithms. A notable weakness of most stability-based generalization bounds is that they hold only in expectation. Generalization with high probability has been established in a landmark paper of Bousque…
Optimal a priori estimates are derived for the population risk, also known as the generalization error, of a regularized residual network model. An important part of the regularized model is the usage of a new path norm, called the weighted path norm, as the regularization term. The weighted path norm treats the skip c…
The success of deep learning has led to a rising interest in the generalization property of the stochastic gradient descent (SGD) method, and stability is one popular approach to study it. Existing works based on stability have studied nonconvex loss functions, but only considered the generalization error of the SGD in…
SGD-trained deep nets have bounds on their generalization error.
problem Bounding generalization error for deep neural networks trained by SGD.
method Combining dynamical control of parameter norms and Rademacher complexity estimates.
result Explicit bounds depend on loss trajectory, work for various architectures.
New methods bound estimation error in high-dimensional statistical problems.
problem Fundamental limits of first order methods in high-dimensional estimation.
method Introduces general first order methods for high-dimensional regression and low-rank matrix estimation.
result Derives optimal lower bounds on estimation error for these methods.
New error bound for diffusion models without dimensionality constraints.
problem Error estimation in diffusion generative models without dimensionality constraints.
method Derive dimension-free error bound using a smooth test functional.
result Explicit, dimension-free bound on generated vs true data distributions.
We derive upper bounds on the generalization error of learning algorithms based on their \emph{algorithmic transport cost}: the expected Wasserstein distance between the output hypothesis and the output hypothesis conditioned on an input example. The bounds provide a novel approach to study the generalization of learni…
The paper tackles deep learning from dependent data, achieving optimal performance.
problem Deep learning from strongly mixing observations, especially with regularization and optimality.
method Sparse-penalized regularization for deep neural networks, oracle inequality for expected excess risk.
result Deep neural network estimator achieves minimax optimal rate for nonparametric autoregression.
Unified learning bound for covariate and concept shifts.
problem Generalization under distribution shift in machine learning.
method Support-agnostic definitions of covariate and concept shifts using entropic optimal transport, leading to a unified error bound applicable to various loss functions and label spaces.
result Development of estimators for shifts with concentration guarantees and the DataShifts algorithm for quantifying and estimating the error bound.
Optimal multiclass U-calibration error found to be Θ(√KT).
problem Online multiclass U-calibration with low regret for all bounded proper losses.
method Follow-the-Perturbed-Leader algorithm and lower bound construction.
result Optimal U-calibration error is Θ(√KT).
Study shows how classifiers can approach Bayes error in high-dimensional settings.
problem Generalization error in high-dimensional perceptrons.
method Proved a formula for generalization error using convex optimization and observed that logistic and hinge regression can approach Bayes error closely.
result Logistic and hinge regression can approach Bayes-optimal generalization error closely in high-dimensional settings.
Paper introduces new bounds linking data compressibility to generalization error.
problem Establishing data-dependent generalization bounds.
method Variable-size compressibility framework linking generalization error to compression rate of input data.
result New bounds depend on empirical data measure, subsuming existing PAC-Bayes and intrinsic dimension bounds.
New method improves solving combinatorial optimization problems with smoothed policies.
problem Solving combinatorial optimization problems repeatedly with varying instances.
method Smoothed policies with controlled random perturbations to linear oracle, leading to differentiable surrogate risk.
result Generalization bound decomposes excess risk into bias, estimation, and optimization components.
New bounds show linear predictors rarely overfit with certain optimization methods.
problem Bounding test error for linear predictors with stochastic optimization methods.
method Coupling argument for fixed point methods like stochastic and batch mirror descent.
result Locally-adapted rates that depend on predictor properties, not global problem structure.
Optimal Gaussian noise mechanisms achieve nearly optimal error in unbiased mean estimation.
problem Efficiently estimating the mean of high-dimensional data while preserving privacy.
method Differential privacy mechanisms with Gaussian noise, focusing on optimal covariance.
result Gaussian noise mechanisms achieve nearly optimal error among all private unbiased mean estimation mechanisms.
The paper connects three machine learning methods to reduce generalization errors.
problem Reducing generalization errors in machine learning models.
method Distributionally robust optimization, Bayesian methods, and regularization.
result Machine learning models can be characterized using distributional uncertainty and robustness measures.
GANs learn distributions well from samples, with rates depending on intrinsic dimension.
problem Learning distributions from samples using GANs.
method Oracle inequality, Hölder functions approximation, neural network approximation, integral probability metrics.
result Convergence rates of GANs depend on intrinsic dimension, not ambient dimension.
Estimates convex hulls of smooth function images with error bounds.
problem Estimating the convex hull of the image of a smooth boundary set.
method Using submersion properties and sampling inputs, derive bounds on Hausdorff distance.
result New tighter and more general error bounds for geometric inference.
The paper provides mean-square error bounds for stochastic approximation algorithms.
problem Error bounds for recursive equations with Markovian disturbances.
method Analysis of mean-square error for stochastic approximation algorithms.
result Mean-square error achieves the optimal rate of O(1/n) under certain conditions. Optimal function approximation with Relu neural networks achieves minimal error.
problem Finding the minimal error in approximating convex functions with Relu networks.
method Established necessary and sufficient conditions for optimal approximations, presented neural network architectures, and proposed an algorithm for convergence.
result Proved the convergence of the proposed algorithm and validated it with experimental results.
New bounds predict deep learning generalization better than existing methods.
problem Predicting generalization errors for deep learning models.
method Function-based PAC-Bayesian bounds that meet multiple desiderata.
result The new bound performs significantly better than existing parameter-based PAC-Bayes bounds.
Paper establishes universal lower bounds and optimal rates for clustering sub-exponential mixture models.
problem Achieving optimal error rates in clustering sub-exponential mixture models.
method Establishes universal lower bounds and demonstrates iterative algorithms' optimality in sub-exponential mixture models.
result Iterative algorithms achieve the universal lower bound in sub-exponential mixture models.
The seminal paper of Caponnetto and de Vito (2007) provides minimax-optimal rates for kernel ridge regression in a very general setting. Its proof, however, contains an error in its bound on the effective dimensionality. In this note, we explain the mistake, provide a correct bound, and show that the main theorem remai…
Score-based diffusion models achieve optimal error bounds under non-parametric assumptions.
problem Improving the minimax optimality of score-based diffusion models.
method Kernel-based score estimation and early stopping strategy.
result Achieves minimax optimal error bounds under sub-Gaussian and Sobolev space assumptions.
The paper develops a minimax optimal method for high-dimensional regression using auxiliary data.
problem High-dimensional additive regression with heavy-tailed errors and transfer learning.
method Smooth backfitting estimator with local linear smoothing, followed by a two-stage estimation method.
result The method achieves the minimax optimal rate under certain conditions.
The paper improves error bounds for Bayesian quadrature in noisy settings.
problem Improving error bounds for Bayesian quadrature in noisy settings.
method Develops a two-step meta-algorithm to relate average-case quadrature error to L2-function approximation error. result Provides new average-case results for various kernels and noise settings.
Optimizes classification algorithms with bounds on error rates.
problem Bounding uncertainties in classifier outputs for diagnostic testing.
method Set-theoretic and probabilistic arguments to derive uniform error bounds.
result Optimal partition minimizes the largest Gershgorin radius of the confusion matrix.
Optimizes pruning masks for neural networks using probabilistic fine-tuning and PAC-Bayes bounds.
problem Improving neural network performance through adaptive pruning of weights.
method Optimizes stochastic pruning masks by minimizing expected loss, considering data-adaptive regularization and feature alignment.
result Probabilistic fine-tuning leads to improved test error over baseline methods in neural networks.
Improved generalization bounds for SGD in non-convex learning.
problem Understanding generalization properties of SGD in non-convex settings.
method Introducing Type II perturbed SGD (T2pm-SGD) to analyze generalization error bounds.
result Tighter generalization error bounds for SGD in non-convex learning, especially for sub-Gaussian and bounded loss functions.