Mutual Information Maximization for Simple and Accurate Part-Of-Speech Induction

ACL 2019

Mutual Information Maximization for Simple and Accurate Part-Of-Speech Induction

Jan 19, 2021
|
36 views
|
Details
Abstract: We address part-of-speech (POS) induction by maximizing the mutual information between the induced label and its context. We focus on two training objectives that are amenable to stochastic gradient descent (SGD): a novel generalization of the classical Brown clustering objective and a recently proposed variational lower bound. While both objectives are subject to noise in gradient updates, we show through analysis and experiments that the variational lower bound is robust whereas the generalized Brown objective is vulnerable. We obtain competitive performance on a multitude of datasets and languages with a simple architecture that encodes morphology and context. Authors: Karl Stratos (Toyota Technological Institute at Chicago)

Comments
loading...