Implicitly constrained semi-supervised linear discriminant analysis

TY - GEN

T1 - Implicitly constrained semi-supervised linear discriminant analysis

AU - Krijthe, Jesse H.

AU - Loog, Marco

N1 - Conference code: 22

PY - 2014/12/4

Y1 - 2014/12/4

N2 - Semi-supervised learning is an important and active topic of research in pattern recognition. For classification using linear discriminant analysis specifically, several semi-supervised variants have been proposed. Using any one of these methods is not guaranteed to outperform the supervised classifier which does not take the additional unlabeled data into account. In this work we compare traditional Expectation Maximization type approaches for semi-supervised linear discriminant analysis with approaches based on intrinsic constraints and propose a new principled approach for semi-supervised linear discriminant analysis, using so-called implicit constraints. We explore the relationships between these methods and consider the question if and in what sense we can expect improvement in performance over the supervised procedure. The constraint based approaches are more robust to misspecification of the model, and may outperform alternatives that make more assumptions on the data in terms of the log-likelihood of unseen objects.

AB - Semi-supervised learning is an important and active topic of research in pattern recognition. For classification using linear discriminant analysis specifically, several semi-supervised variants have been proposed. Using any one of these methods is not guaranteed to outperform the supervised classifier which does not take the additional unlabeled data into account. In this work we compare traditional Expectation Maximization type approaches for semi-supervised linear discriminant analysis with approaches based on intrinsic constraints and propose a new principled approach for semi-supervised linear discriminant analysis, using so-called implicit constraints. We explore the relationships between these methods and consider the question if and in what sense we can expect improvement in performance over the supervised procedure. The constraint based approaches are more robust to misspecification of the model, and may outperform alternatives that make more assumptions on the data in terms of the log-likelihood of unseen objects.

U2 - 10.1109/ICPR.2014.646

DO - 10.1109/ICPR.2014.646

M3 - Article in proceedings

SP - 3762

EP - 3767

BT - 2014 22nd International Conference on Pattern Recognition (ICPR)

PB - IEEE

T2 - 22nd International Conference on Pattern Recognition

Y2 - 24 August 2014 through 28 August 2014

ER -