Experiments | Next we used collapsed Gibbs sampling to infer a distribution over topics, 6?, for each of the relations in the primary corpus (based solely on tuples in the training set) using the topics from the generalization corpus. |
Experiments | To evaluate how well our topic-class associations carry over to unseen relations we used the same random sample of 100 relations from the pseudo-disambiguation experiment.8 For each argument of each relation we picked the top two topics according to frequency in the 5 Gibbs samples . |
Previous Work | Additionally we perform full Bayesian inference using collapsed Gibbs sampling , in which parameters are integrated out (Griffiths and Steyvers, 2004). |
Topic Models for Selectional Prefs. | For all the models we use collapsed Gibbs sampling for inference in which each of the hidden variables (e. g., 27.,“ and 273,32 in LinkLDA) are sampled sequentially conditioned on a full-assignment to all others, integrating out the parameters (Griffiths and Steyvers, 2004). |
Topic Models for Selectional Prefs. | In addition, there are several scalability enhancements such as SparseLDA (Yao et al., 2009), and an approximation of the Gibbs Sampling procedure can be efficiently parallelized (Newman et al., 2009). |
Abstract | We formalize nonparametric Bayesian STSG with epsilon alignment in full generality, and provide a Gibbs sampling algorithm for posterior inference tailored to the task of extractive sentence compression. |
Evaluation | We compared the Gibbs sampling compressor (GS) against a version of maximum a posteriori EM (with Dirichlet parameter greater than 1) and a discriminative STSG based on SVM training (Cohn and Lapata, 2008) (SVM). |
The STSG Model | 3.2 Posterior inference via Gibbs sampling |
The STSG Model | We use Gibbs sampling (Geman and Geman, 1984), a Markov chain Monte Carlo (MCMC) method, to sample from the posterior (3). |
Related work | The combination of a well-defined probabilistic model and Gibbs sampling procedure for estimation guarantee (eventual) convergence and the avoidance of degenerate solutions. |
Three selectional preference models | Following Griffiths and Steyvers (2004), we estimate the model by Gibbs sampling . |
Three selectional preference models | As suggested by the similarity between (4) and (2), the ROOTH-LDA model can be estimated by an LDA-like Gibbs sampling procedure. |