We study the impact of big models (in terms of the degree of lexicalization) and big data (in terms of the training corpus size) on dependency grammar induction. We experimented with L-DMV, a lexicalized version of Dependency Model with Valence and L-NDMV, our lexicalized extension of the Neural Dependency Model with Valence. We find that L-DMV only benefits from very small degrees of lexicalization and moderate sizes of training corpora. L-NDMV can benefit from big training data and lexicalization of greater degrees, especially when enhanced with good model initialization, and it achieves a result that is competitive with the current state-of-the-art.
@article{arxiv.1708.00801,
title = {Dependency Grammar Induction with Neural Lexicalization and Big Training Data},
author = {Wenjuan Han and Yong Jiang and Kewei Tu},
journal= {arXiv preprint arXiv:1708.00801},
year = {2017}
}