Unsupervised Domain Adaptation with Feature Embeddings
Computation and Language
2015-04-17 v3 Machine Learning
Abstract
Representation learning is the dominant technique for unsupervised domain adaptation, but existing approaches often require the specification of "pivot features" that generalize across domains, which are selected by task-specific heuristics. We show that a novel but simple feature embedding approach provides better performance, by exploiting the feature template structure common in NLP problems.
Cite
@article{arxiv.1412.4385,
title = {Unsupervised Domain Adaptation with Feature Embeddings},
author = {Yi Yang and Jacob Eisenstein},
journal= {arXiv preprint arXiv:1412.4385},
year = {2015}
}
Comments
For more details, please refer to the long version of this paper: http://www.cc.gatech.edu/~jeisenst/papers/yang-naacl-2015.pdf