erasmus

At GE, we rise to the challenge of developing a globe that performs. Subject models give a helpful strategy for dimensionality reduction and exploratory data analysis in significant text corpora. Most approaches to subject model inference have been primarily based on a maximum likelihood objective. Efficient algorithms exist that approximate this objective, but they have no provable guarantees. Not too long ago, algorithms have been introduced that supply provable bounds, but these algorithms are not practical simply because they are inefficient and not robust to violations of model assumptions. In this paper we present an algorithm for subject model inference...