Learning from non-iid data: fast rates for the one-vs-all multiclass plug-in classifiers

DOI10.1007/978-3-319-17142-5_32zbMATH Open1459.68169arXiv1408.2714OpenAlexW311675162MaRDI QIDQ2948481FDOQ2948481

Authors:

Publication date: 30 September 2015

Published in: Lecture Notes in Computer Science (Search for Journal in Brave)

Abstract: We prove new fast learning rates for the one-vs-all multiclass plug-in classifiers trained either from exponentially strongly mixing data or from data generated by a converging drifting distribution. These are two typical scenarios where training data are not iid. The learning rates are obtained under a multiclass version of Tsybakov's margin assumption, a type of low-noise assumption, and do not depend on the number of classes. Our results are general and include a previous result for binary-class plug-in classifiers with iid data as a special case. In contrast to previous works for least squares SVMs under the binary-class setting, our results retain the optimal learning rate in the iid case.

Full work available at URL: https://arxiv.org/abs/1408.2714

Recommendations

• Fast learning rates for plug-in classifiers
• On the Rate of Convergence of Local Averaging Plug-In Classification Rules Under a Margin Condition
• Fast classification rates without standard margin assumptions
• Classification with non-i.i.d. sampling
• Fast rates in statistical and online learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• In defense of one-vs-all classification
• Prediction of time series by statistical learning: general losses and fast rates
• Analysis of comparative data with hierarchical autocorrelation
• Exponential inequalities for sums of random vectors
• Optimal aggregation of classifiers in statistical learning.
• Fast learning rates for plug-in classifiers
• Fast rates for support vector machines using Gaussian kernels
• Minimum complexity regression estimation with weakly dependent observations
• New analysis and algorithm for learning with drifting distributions
• On the Rate of Convergence of Local Averaging Plug-In Classification Rules Under a Margin Condition
• Statistical analysis of some multi-category large margin classification methods
• Generalization error for multi-class margin classification
• Classification via local multi-resolution projections
• On the complexity of learning from drifting distributions
• The complexity of learning according to two models of a drifting environment
• Plug-in Approach to Active Learning
• Fast learning from \(\alpha\)-mixing observations
• Generalization and Robustness of Batched Weighted Average Algorithm with V-Geometrically Ergodic Markov Data

Cited In (7)

• Recovery guarantees for polynomial coefficients from weakly dependent data with outliers
• Adaptive group Lasso neural network models for functions of few variables and time-dependent data
• Fast learning rates for plug-in classifiers
• Posterior concentration and fast convergence rates for generalized Bayesian learning
• An adaptive multiclass nearest neighbor classifier
• A generalization bound of deep neural networks for dependent data
• Towards convergence rate analysis of random forests for classification

This page was built for publication: Learning from non-iid data: fast rates for the one-vs-all multiclass plug-in classifiers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2948481)