Unsupervised Latent Tree Induction with Deep Inside-Outside Recursive Autoencoders
2019-04-03Code Available1· sign in to hype
Andrew Drozdov, Pat Verga, Mohit Yadav, Mohit Iyyer, Andrew McCallum
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/iesl/dioraOfficialIn paperpytorch★ 0
- github.com/i-machine-think/emergent_grammar_inductionnone★ 11
- github.com/rgalhama/diorapytorch★ 0
Abstract
We introduce deep inside-outside recursive autoencoders (DIORA), a fully-unsupervised method for discovering syntax that simultaneously learns representations for constituents within the induced tree. Our approach predicts each word in an input sentence conditioned on the rest of the sentence and uses inside-outside dynamic programming to consider all possible binary trees over the sentence. At test time the CKY algorithm extracts the highest scoring parse. DIORA achieves a new state-of-the-art F1 in unsupervised binary constituency parsing (unlabeled) in two benchmark datasets, WSJ and MultiNLI.