The Infinite PCFG Using Hierarchical Dirichlet Processes

By Percy Liang, Slav Petrov, Michael I. Jordan, Daniel Klein. 2007

225 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Variational Inference: A Review for Statisticians (2016, methodology): “In natu-ral language processing, variational inference has been used for solving problems such as parsing (Liang et al. 2007; Liang, Jordan, and Klein 2009), grammar induction (Kurihara and Sato 2006; Naseem et al. 2010; Cohen and Smith 2010), models of streaming text (Yogatama et al. 2014), topic…”
The nested chinese restaurant process and bayesian nonparametric inference of topic hierarchies (2007, background): “2007], and natural language processing [Liet al. 2007; Teh et al. 2007b; Goldwater et al. 2006a, 2006b; Johnson et al. 2007;Liang et al. 2007].”
Bayesian Nonparametrics: Hierarchical Bayesian nonparametric models with applications (2010, methodology): “As a further generalization of the HDP-HMM, several groups have considered nonparametric versions of probabilistic grammars (Johnson et al., 2007; Liang et al., 2007; Finkel et al., 2007).”
Unsupervised Lexicon Discovery from Acoustic Input (2015, methodology): “Note that to simultaneously infer a phonetic inventory of an unknown size and model phonetic variation, we can use the infinite PCFG (Liang et al., 2007) to”
Compound Probabilistic Context-Free Grammars for Grammar Induction (2019, background): “…Manning, 2002), priors or Code: https://github.com/harvardnlp/compound-pcfg non-parametric models (Kurihara and Sato, 2006; Johnson et al., 2007; Liang et al., 2007; Wang and Blunsom, 2013), and manually-engineered features (Huang et al., 2012; Golland et al., 2012) to encourage the desired…”
Covariance in Unsupervised Learning of Probabilistic Grammars (2010, methodology): “, 2009) as well as over the states in a PCFG (Finkel et al., 2007; Liang et al., 2007), and even over grammatical derivation structures larger than context-free production rules (Johnson et al.”
Self-Training with Products of Latent Variable Grammars (2010, background): “However, because the latent variable grammars are not explicitly regularized, EM keeps ﬁt-ting the training data and eventually begins over-ﬁtting (Liang et al., 2007).”
Syntactic Topic Models (2008, background): “(1) Following [16], q(β|β∗) is not a full distribution, but is a degenerate point estimate truncated so that all weights whose index is greater than K are zero in the variational distribution.”
Online Variational Inference for the Hierarchical Dirichlet Process (2011, methodology): “As a consequence, it is not amendable to closed form variational updates [3, 4].”
Modeling Human Understanding of Complex Intentional Action with a Bayesian Nonparametric Subgoal Model (2015, methodology): “To enhance the expressiveness of our model, Infinite PCFGs (Liang et al. 2007), Adaptor Grammars (Johnson, Griffiths, and Goldwater 2007), or fragment grammars (O’Donnell 2015) are promising extensions to the simple Dirichlet Process we employ.”
Inducing Tree-Substitution Grammars (2010, methodology): “…grammar induction, which we will refer to as grammar refinement because the learned grammars can be viewed as refinements of the treebank PCFG, include the symbol-splitting approach of Petrov and Klein (2007) and the tree-substitution grammars of Data-Oriented Parsing (Bod et al., 2003; Bod, 2003).”
Incremental Sigmoid Belief Networks for Grammar Learning (2010, methodology): “To solve this problem the splitand-merge approach was considered in (Petrov et al., 2006; Petrov and Klein, 2007) and Dirichlet Process priors in Liang et al. (2007).”
A Latent Variable Model for Generative Dependency Parsing (2007, background): “In [28, 41, 39, 25, 40] non-terminals in a standard PCFG model are augmented with latent variables.”
Bayesian Nonparametric Poisson Factorization for Recommendation Systems (2014, methodology): “, q(vuk) = δτuk(vuk), an alternative that is widely used in the BNP literature (Liang et al., 2007; Bryant & Sudderth, 2012).”
Using Universal Linguistic Knowledge to Guide Grammar Induction (2010, methodology): “While most of the existing approaches are implemented in the supervised setting (Finkel et al., 2007; Petrov and Klein, 2007), Liang et al. (2007) propose a non-parametric Bayesian model that learns the granularity of PCFG categories in an unsupervised fashion.”