Thomas Hofmann
Thomas Hofmann studies statistical learning and text. His probabilistic latent semantic analysis paper models co-occurrence data with latent classes. Hofmann’s UAI 1999 paper presents a latent-class mixture model for co-occurrence data. The paper applies the model to information retrieval and learning from text, and uses a tempered expectation-maximization method to control overfitting. He joined the Brown faculty in 1999, received tenure in 2004, then moved to a professorship at Darmstadt and led the IPSI institute. He joined Google in Zurich as an engineering director in 2006. He left Google in early 2014 and became an ETH computer-science professor in April 2014.
- Now at
- ETH Zurich
- Works
- 1,908
Publication numbers from OpenAlex, as of 14 October 2026. Citations and h-index are hidden because this OpenAlex record may be incomplete or mix in other authors.
Timeline
-
1999
Probabilistic latent semantic analysis
Hofmann’s UAI 1999 paper presents a latent-class mixture model for co-occurrence data. The paper applies the model to information retrieval and learning from text, and uses a tempered expectation-maximization method to control overfitting.
-
1999–2004
University research
He joined the Brown faculty in 1999, received tenure in 2004, then moved to a professorship at Darmstadt and led the IPSI institute.
-
2006–2014
Google and an academic return
He joined Google in Zurich as an engineering director in 2006. He left Google in early 2014 and became an ETH computer-science professor in April 2014.
Sources and credits
Photo: ethar.ethz.ch; creator not stated. License: Not stated by source. Source: ethar.ethz.ch. Cropped to a square and resized. Ask us to remove or replace this photo.