University of Cambridge > Talks.cam > Inference Group > Topic Models -- and how to break them

  Browse A-Z

Log in

University Account External (via Google)

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Topic Models -- and how to break them

Download to your calendar using vCal

If you have a question about this talk, please contact Philip Sterne .

In the first half of this talk, I will given an introduction to topic models (mixtures of discrete distributions). I will also develop several models allowing inference conditional on features (metadata) of the document.

In a second part, I will present some negative results from an attempt to apply such models to very short documents from the web, demonstrating the limits of these models.

This talk is part of the Inference Group series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Maths Knowledge Transfer Seminar room 1, Hopkins Building, Department of Biochemistry, Downing Site, University of Cambridge CB2 1QW Part III Seminar Series Lent 2008

Other talks

Synthesizing Diverse Policies for Multi-Agent Coordination Classical Reductions and Converse Bounds in Quantum Hypothesis Testing Reversals of the Geodynamo Title to be confirmed CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models Title TBC
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity