Local Rademacher complexities
From MaRDI portal
Publication:2583411
Abstract: We propose new bounds on the error of learning algorithms in terms of a data-dependent notion of complexity. The estimates we establish give optimal rates and are based on a local and empirical version of Rademacher averages, in the sense that the Rademacher averages are computed from the data, on a subset of functions with small empirical error. We present some applications to classification and prediction with convex function classes, and with kernel classes in particular.
Recommendations
Cites work
- scientific article; zbMATH DE number 2089352 (Why is no real title available?)
- scientific article; zbMATH DE number 2089354 (Why is no real title available?)
- scientific article; zbMATH DE number 5654889 (Why is no real title available?)
- scientific article; zbMATH DE number 49190 (Why is no real title available?)
- scientific article; zbMATH DE number 1246230 (Why is no real title available?)
- scientific article; zbMATH DE number 2034517 (Why is no real title available?)
- scientific article; zbMATH DE number 1552503 (Why is no real title available?)
- scientific article; zbMATH DE number 893887 (Why is no real title available?)
- 10.1162/153244303321897690
- A Bennett concentration inequality and its application to suprema of empirical processes
- A distribution-free theory of nonparametric regression
- A new approach to least-squares estimation, with applications
- A sharp concentration inequality with applications
- About the constants in Talagrand's concentration inequalities for empirical processes.
- Advanced lectures on machine learning. Machine learning summer school 2002, Canberra, Australia, February 11--22, 2002. Revised lectures
- Asymptotic Statistics
- Complexity regularization via localized random penalties
- Concentration inequalities using the entropy method
- Convergence of stochastic processes
- Convexity, Classification, and Risk Bounds
- Decision theoretic generalizations of the PAC model for neural net and other learning applications
- Empirical minimization
- Improving the sample complexity using global data
- Model selection and error estimation
- On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities
- Rademacher averages and phase transitions in Glivenko-Cantelli classes
- Rademacher penalties and structural risk minimization
- Sharper bounds for Gaussian and empirical processes
- Smooth discrimination analysis
- Some applications of concentration inequalities to statistics
- Sphere packing numbers for subsets of the Boolean \(n\)-cube with bounded Vapnik-Chervonenkis dimension
- The importance of convexity in learning with squared loss
- Une inégalité de Bennett pour les maxima de processus empiriques. (A Bennet type inequality for maxima of empirical processes)
- Uniform Central Limit Theorems
- Weak convergence and empirical processes. With applications to statistics
Cited in
(only showing first 100 items - show all)- Concentration inequalities for samples without replacement
- Learning models with uniform performance via distributionally robust optimization
- A tight upper bound on the generalization error of feedforward neural networks
- Full error analysis for the training of deep neural networks
- Online pairwise learning algorithms with convex loss functions
- Online regularized learning with pairwise loss functions
- Bayesian fractional posteriors
- A reproducing kernel Hilbert space approach to high dimensional partially varying coefficient model
- Rademacher penalties and structural risk minimization
- Rademacher averages and phase transitions in Glivenko-Cantelli classes
- Statistics of robust optimization: a generalized empirical likelihood approach
- Fast learning rate of non-sparse multiple kernel learning and optimal regularization strategies
- Localized Gaussian width of \(M\)-convex hulls with applications to Lasso and convex aggregation
- Guaranteed Functional Tensor Singular Value Decomposition
- Calibration of \(\epsilon\)-insensitive loss in support vector machines regression
- Penalized empirical risk minimization over Besov spaces
- On the empirical estimation of integral probability metrics
- Suboptimality of constrained least squares and improvements via non-linear predictors
- Orthogonal statistical learning
- Local learning estimates by integral operators
- Non-convex projected gradient descent for generalized low-rank tensor regression
- PAC-learning with approximate predictors
- scientific article; zbMATH DE number 7370593 (Why is no real title available?)
- scientific article; zbMATH DE number 7307489 (Why is no real title available?)
- Estimates of the approximation error using Rademacher complexity: Learning vector-valued functions
- Rademacher Margin Complexity
- Generalization bounds for non-stationary mixing processes
- Model selection in reinforcement learning
- Robustness and generalization
- Estimating individualized treatment rules using outcome weighted learning
- Combinatorial bounds for learning performance
- Metamodel construction for sensitivity analysis
- Convergence rates for empirical barycenters in metric spaces: curvature, convexity and extendable geodesics
- Transfer bounds for linear feature learning
- On the optimal estimation of probability measures in weak and strong topologies
- Consistency analysis of an empirical minimum error entropy algorithm
- Tikhonov, Ivanov and Morozov regularization for support vector machine learning
- Sample average approximation with heavier tails. I: Non-asymptotic bounds with weak assumptions and stochastic constraints
- Oracle inequalities for sparse additive quantile regression in reproducing kernel Hilbert space
- Learning rates for partially linear support vector machine in high dimensions
- Bootstrap model selection for possibly dependent and heterogeneous data
- Multi-kernel learning for multi-label classification with local Rademacher complexity
- VC dimension, fat-shattering dimension, Rademacher averages, and their applications
- Kernelized elastic net regularization: generalization bounds, and sparse recovery
- Model selection by resampling penalization
- Sample average approximation with heavier tails II: localization in stochastic convex optimization and persistence results for the Lasso
- An improved analysis of the Rademacher data-dependent bound using its self bounding property
- Local Rademacher complexity-based learning guarantees for multi-task learning
- Empirical minimization
- Optimal prediction for high-dimensional functional quantile regression in reproducing kernel Hilbert spaces
- scientific article; zbMATH DE number 2089353 (Why is no real title available?)
- Singularity, misspecification and the convergence rate of EM
- Convergence rates of least squares regression estimators with heavy-tailed errors
- Analysis of the generalization error: empirical risk minimization over deep artificial neural networks overcomes the curse of dimensionality in the numerical approximation of Black-Scholes partial differential equations
- scientific article; zbMATH DE number 1804106 (Why is no real title available?)
- Handling concept drift via model reuse
- Robust multicategory support vector machines using difference convex algorithm
- Using the doubling dimension to analyze the generalization of learning algorithms
- Influence diagnostics in support vector machines
- An error analysis for deep binary classification with sigmoid loss
- Margin-adaptive model selection in statistical learning
- Comments on: Support vector machines maximizing geometric margins for multi-class classification
- Convergence rates of generalization errors for margin-based classification
- Distribution-dependent sample complexity of large margin learning
- ERM and RERM are optimal estimators for regression problems when malicious outliers corrupt the labels
- Estimation bounds and sharp oracle inequalities of regularized procedures with Lipschitz loss functions
- Graphical convergence of subgradients in nonconvex optimization and learning
- scientific article; zbMATH DE number 2089354 (Why is no real title available?)
- scientific article; zbMATH DE number 7625158 (Why is no real title available?)
- Distribution-free robust linear regression
- Regularization and the small-ball method. II: Complexity dependent error rates
- Boosting with early stopping: convergence and consistency
- Empirical variance minimization with applications in variance reduction and optimal control
- Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder)
- A local Vapnik-Chervonenkis complexity
- Estimating conditional quantiles with the help of the pinball loss
- Smooth sparse coding via marginal regression for learning sparse representations
- Deep nonlinear sufficient dimension reduction
- Locally simultaneous inference
- scientific article; zbMATH DE number 7370625 (Why is no real title available?)
- Improvement of multiple kernel learning using adaptively weighted regularization
- An elementary analysis of ridge regression with random design
- Rademacher complexity for Markov chains: applications to kernel smoothing and Metropolis-Hastings
- Multi-kernel regularized classifiers
- Data-adaptive discriminative feature localization with statistically guaranteed interpretation
- Minimax rates for conditional density estimation via empirical entropy
- Compressive statistical learning with random feature moments
- Fast rates for support vector machines using Gaussian kernels
- Strong overall error analysis for the training of artificial neural networks via random initializations
- Newton Sketch: A Near Linear-Time Optimization Algorithm with Linear-Quadratic Convergence
- Overall error analysis for the training of deep neural networks via stochastic gradient descent with random initialisation
- Researches on Rademacher complexities in statistical learning theory: a survey
- A unified penalized method for sparse additive quantile models: an RKHS approach
- Continuity of Performance Metrics for Thin Feature Maps
- Rademacher complexity and grammar induction algorithms: what it may (not) tell us
- Benign Overfitting and Noisy Features
- Complexity regularization via localized random penalties
- Approximation properties of certain operator-induced norms on Hilbert spaces
- Noisy discriminant analysis with boundary assumptions
- Complexity control in statistical learning
This page was built for publication: Local Rademacher complexities
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2583411)
