We present a shared logistic normal distribution as a Bayesian prior over probabilistic grammar weights. This approach generalizes the similar use of logistic normal distributions [3], enabling soft parameter tying during inference across different multinomials comprising the probabilistic grammar. We show that this model outperforms previous approaches on an unsupervised dependency grammar induction task.
We present a family of priors over probabilistic grammar weights, called the shared logistic normal distribution. This family extends the partitioned logistic normal distribution, enabling factored covariance between the probabilities of different derivation events in the probabi
In this paper some properties of logistics - x family are discussed and a member of the family, the logistic–normal distribution, is studied in detail. Average deviations, risk function and fashion for logistic–normal distribution is obtained. The method of maximum l
We present the discrete infinite logistic normal distribution (DILN, “Dylan”), a Bayesian nonparametric prior for mixed membership models. DILN is a generalization of the hierarchical Dirichlet process (HDP) that models correlation structure between the weights of the atoms at th
We present the discrete infinite logistic normal distribution (DILN), a Bayesian nonparametric prior for mixed membership models. DILN generalizes the hierarchical Dirichlet process (HDP) to model correlation structure between the weights of the atoms at the group level. We deriv
Mixed-membership models (e.g. “topic models”) are inarguably popular; especially latent Dirichlet allocation (LDA) [Blei et al., 2003] and its variants. Such models have become a fundamental tool in the analysis and exploration of many types of data. Originally designed to model