Most large topic models are approximately separable

@article{Ding2015MostLT,
  title={Most large topic models are approximately separable},
  author={Weicong Ding and Prakash Ishwar and Venkatesh Saligrama},
  journal={2015 Information Theory and Applications Workshop (ITA)},
  year={2015},
  pages={199-203}
}
Separability has recently been leveraged as a key structural condition in topic models to develop asymptotically consistent algorithms with polynomial statistical and computational efficiency guarantees. Separability corresponds to the presence of at least one novel word for each topic. Empirical estimates of topic matrices for Latent Dirichlet Allocation models have been observed to be approximately separable. Separability may be a convenient structural property, but it appears to be too… CONTINUE READING

From This Paper

Figures, tables, and topics from this paper.

Similar Papers

Loading similar papers…