On semi-supervised learning
This work addresses when and why semi-supervised learning works for classification problems with limited labeled data, but it is incremental as it builds on existing algorithms and focuses on well-conditioned scenarios.
The paper tackles the problem of semi-supervised learning by proposing a new algorithm that, under necessary conditions, asymptotically achieves the performance of the best theoretical rule as unlabeled data increases, with performance assessed on the Isolet dataset showing strong dependence on initial training samples.
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.