ordinal embedding

Thursday, May 19, 2016 - 9:00am - 9:50am
Robert Nowak (University of Wisconsin, Madison)
Ordinal embedding, also known as non-metric multidimensional scaling, aims to represent items as points in d-dimensional Euclidean space so that the distances between points agree as well as possible with a given set of ordinal comparisons such as item i is closer to item j than to item k. This classic problem is often used to visualize perceptual similarities. Recently, several authors have proposed a variety of algorithms for learning a metric embedding from ordinal information.
Subscribe to RSS - ordinal embedding