On semi-supervised learning
From MaRDI portal
Publication:2665783
Abstract: Semi-supervised learning deals with the problem of how, if possible, to take advantage of a huge amount of unclassified data, to perform a classification in situations when, typically, there is little labeled data. Even though this is not always possible (it depends on how useful, for inferring the labels, it would be to know the distribution of the unlabeled data), several algorithm have been proposed recently. %but in general they are not proved to outperform A new algorithm is proposed, that under almost necessary conditions, %and it is proved that it attains asymptotically the performance of the best theoretical rule as the amount of unlabeled data tends to infinity. The set of necessary assumptions, although reasonable, show that semi-supervised classification only works for very well conditioned problems. The focus is on understanding when and why semi-supervised learning works when the size of the initial training sample remains fixed and the asymptotic is on the size of the unlabeled data. The performance of the algorithm is assessed in the well known "Isolet" real-data of phonemes, where a strong dependence on the choice of the initial training sample is shown.
Recommendations
- Generalization error bounds in semi-supervised classification under the cluster assumption
- A discriminative model for semi-supervised learning
- Semi-supervised learning via constraints
- On the choice of kernel and labelled data in semi-supervised learning methods
- Density-sensitive semisupervised inference
Cites work
- A plug-in approach to support estimation
- Curvature Measures
- Density-sensitive semisupervised inference
- Detection of low dimensionality and data denoising via set estimation techniques
- Generalization error bounds in semi-supervised classification under the cluster assumption
- scientific article; zbMATH DE number 1332320 (Why is no real title available?)
- Learning with a probabilistic teacher
- Manifold regularization and semi-supervised learning: some theoretical analyses
- On boundary estimation
- On Poincaré cone property
- On statistical properties of sets fulfilling rolling-type conditions
- On the strong uniform consistency of a new kernel density estimator
- On transductive support vector machines
- Probability of error of some adaptive pattern-recognition machines
- Semi-supervised learning on Riemannian manifolds
- Some remarks on the measurability of certain sets
- The relative value of labeled and unlabeled samples in pattern recognition with an unknown mixing parameter
Cited in
(46)- Analysis of presence-only data via semi-supervised learning approaches
- Projected estimators for robust semi-supervised classification
- Semi-supervised learning based on high density region estimation
- Combining labelled and unlabelled data in the design of pattern classification systems
- Minimax semi-supervised set-valued approach to multi-class classification
- Margin-based semi-supervised learning using Apollonius circle
- An apparent paradox: a classifier based on a partially classified sample may have smaller expected error rate than that if the sample were completely classified
- Semi-supervised inference: general theory and estimation of means
- Safe semi-supervised learning based on weighted likelihood
- Asymptotic comparison of semi-supervised and supervised linear discriminant functions for heteroscedastic normal populations
- Online reliable semi-supervised learning on evolving data streams
- On semi-supervised linear regression in covariate shift problems
- Semi-supervised learning using ensembles of multiple 1D-embedding-based label boosting
- On efficient large margin semisupervised learning: method and theory
- Generalized expectation criteria for semi-supervised learning with weakly labeled data
- Unsupervised supervised learning. I: Estimating classification and regression errors without labels
- Improved generative semisupervised learning based on finely grained component-conditional class labeling
- Unbiased generative semi-supervised learning
- On the choice of kernel and labelled data in semi-supervised learning methods
- Unlabeled data does provably help
- Semi-supervised regression and system identification
- Semi-supervised logistic discrimination via regularized Gaussian basis expansions
- Generalization error bounds in semi-supervised classification under the cluster assumption
- scientific article; zbMATH DE number 5957483 (Why is no real title available?)
- Density-sensitive semisupervised inference
- Semi-supervised learning with density-ratio estimation
- scientific article; zbMATH DE number 6129062 (Why is no real title available?)
- Semi-Supervised Approaches to Efficient Evaluation of Model Prediction Performance
- Semi-supervised learning for regression based on the diffusion matrix
- Consistency of plug-in confidence sets for classification in semi-supervised learning
- Semi-supervised learning in causal and anticausal settings
- Semisupervised Learning for Computational Linguistics
- Unsupervised supervised learning. II: Margin-based classification without labels
- scientific article; zbMATH DE number 6253975 (Why is no real title available?)
- Open Problems in Efficient Semi-supervised PAC Learning
- Automatic bias correction methods in semi-supervised learning
- Semi-supervised learning via constraints
- Machine Learning: ECML 2004
- scientific article; zbMATH DE number 2243385 (Why is no real title available?)
- Semi-supervised interpolation in an anticausal learning scenario
- Interpolation consistency training for semi-supervised learning
- Semi-Supervised Linear Regression
- Almost exact recovery in label spreading
- Semi-supervised learning with the help of Parzen windows
- Combining semi-supervised clustering and classification under a generalized framework
- A case-study on naïve labelling for the nearest mean and the linear discriminant classifiers
This page was built for publication: On semi-supervised learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2665783)
