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

EVENT:

Improving performance of topic models by variable grouping
Conferences & Talks

 

description

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 

Additive Manufacturing for Electronics “Beyond Moore”
Janos Veres
5 December 2016 | San Francisco, CA
Conferences & Talks  

Metamaterial Frequency-Adaptive Steerable Technology (M-FAST)
6 December 2016 - 8 December 2016 | London, United Kingdom
Conferences & Talks  

The Future of Work
Stephen Hoover
12 December 2016 | Stanford, CA
Conferences & Talks  

Connecting through Play: Tech to Augment In-Person Interaction
Katherine Isbister
15 December 2016 | George E. Pake Auditorium, PARC
PARC Forum