home › event - improving performance of topic models by variable grouping


Improving performance of topic models by variable grouping
Conferences & Talks



Topic models have a wide range of applications, including modeling of text documents, images, user preferences, product rankings, and many others. However, learning optimal models may be difficult, especially for large problems. The reason is that inference techniques such as Gibbs sampling often converge to suboptimal models due to the abundance of local minima in large datasets. In this paper, we propose a general method of improving the performance of topic models. The method, called 'grouping transform,' works by introducing auxiliary variables which represent assignments of the original model tokens to groups. Using these auxiliary variables, it becomes possible to re-sample an entire group of tokens at a time. This allows the sampler to make larger state space moves. As a result, better models are learned and performance is improved. The proposed ideas are illustrated on several topic models and several text and image datasets. We show that the grouping transform significantly improves performance over standard models.


upcoming events   view all 

What is the Future of Cybersecurity?
Alissa Johnson
26 September 2017 | George E. Pake Auditorium, PARC
PARC Forum  

Bringing Reliable (and Transparent) AI to Business (Keynote)
Tolga Kurtoglu, Keynote Presenter
28 September 2017 | Santa Clara, CA
Conferences & Talks  

AI’s Impact on Industry (Keynote Power Panel)
Tolga Kurtoglu, Moderator
28 September 2017 | Santa Clara, CA
Conferences & Talks