Learning from relevant tasks only

Reference:

Samuel Kaski and Jaakko Peltonen. Learning from relevant tasks only. In Joost N. Kok, Jacek Koronacki, Ramon Lopez de Mantaras, Stan Matwin, Dunja Mladenic, and Andrzej Skowron, editors, Machine Learning: ECML 2007 (Proceedings of the 18th European Conference on Machine Learning), pages 608–615, Berlin, 2007. Springer-Verlag.

Abstract:

We introduce a problem called relevant subtask learning, a variant of multi-task learning. The goal is to build a classifier for a task-of-interest having too little data. We also have data for other tasks but only some are relevant, meaning they contain samples classified in the same way as in the task-of-interest. The problem is how to utilize this ``background data'' to improve the classifier in the task-of-interest. We show how to solve the problem for logistic regression classifiers, and show that the solution works better than a comparable multi-task learning model. The key is to assume that data of all tasks are mixtures of relevant and irrelevant samples, and model the irrelevant part with a sufficiently flexible model such that it does not distort the model of relevant data.

Suggested BibTeX entry:

@inproceedings{Kaski07ecml,
    address = {Berlin},
    author = {Samuel Kaski and Jaakko Peltonen},
    booktitle = {Machine Learning: ECML 2007 (Proceedings of the 18th European Conference on Machine Learning)},
    editor = {Joost N. Kok and Jacek Koronacki and Ramon {Lopez de Mantaras} and Stan Matwin and Dunja Mladenic and Andrzej Skowron},
    pages = {608-615},
    publisher = {Springer-Verlag},
    title = {Learning from relevant tasks only},
    year = {2007},
}