University of Cambridge > Talks.cam > Inference Group > Clustering by linear programming, convex optimization and belief propagation

Clustering by linear programming, convex optimization and belief propagation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact David MacKay.

Note unusual time

A popular approach to clustering is to identify a small set of data points called exemplars, and associate every other data point with an exemplar. The goal is to maximize the sum of similarities between data points and their exemplars. This method can be used to cluster vector-space data, but can also be applied to non-vector and even non-metric data, since all that is needed is a set of similarities between pairs of data points. In fact, data points and exemplars can come from different spaces, eg, the data points could be disaster victims while the exemplars are potential food repositories, or the data points could be regions of space to be imaged while the exemplars are potential telescopes. In this talk, I’ll review the state-of-the-art in algorithms for exemplar-based clustering, including recently-proposed ones based on convex optimization, loopy belief propagation and linear programming. I’ll also present benchmarks for these methods.

This talk is part of the Inference Group series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2020 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity