Cross-Task Knowledge-Constrained Self Training
Machine Learning
2009-07-07 v1 Computation and Language
Abstract
We present an algorithmic framework for learning multiple related tasks. Our framework exploits a form of prior knowledge that relates the output spaces of these tasks. We present PAC learning results that analyze the conditions under which such learning is possible. We present results on learning a shallow parser and named-entity recognition system that exploits our framework, showing consistent improvements over baseline methods.
Cite
@article{arxiv.0907.0784,
title = {Cross-Task Knowledge-Constrained Self Training},
author = {Hal Daumé},
journal= {arXiv preprint arXiv:0907.0784},
year = {2009}
}