Витерби EM
Related Papers
Viterbi Training Improves Unsupervised Dependency Parsing
We show that Viterbi (or “hard”) EM is well-suited to unsupervised grammar in- duction. It is more accurate than standard inside-outside re-estimation (classic EM), significantly faster, and simpler. Our ex- …...
Unambiguity Regularization for Unsupervised Learning of Probabilistic Grammars
We introduce a novel approach named unam- biguity regularization for unsupervised learn- ing of probabilistic natural language gram- mars. The approach is based on the observa- tion that natural language …...
Nonconvex Global Optimization for Latent-Variable Models
Many models in NLP involve latent vari- ables, such as unknown parses, tags, or alignments. Finding the optimal model pa- rameters is then usually a difficult noncon- vex optimization problem. …...
Generalizing a Strongly Lexicalized Parser using Unlabeled Data
Statistical parsers trained on labeled data suffer from sparsity, both grammatical and lexical. For parsers based on strongly lexicalized grammar formalisms (such as CCG, which has complex lexical cate- gories …...
Unsupervised Discourse Constituency Parsing Using Viterbi EM
AbstractIn this paper, we introduce an unsupervised discourse constituency parsing algorithm. We use Viterbi EM with a margin-based criterion to train a span-based discourse parser in an unsupervised manner. We …...