UMAP for the working mathematician

Abstract: The Healy-McInnes UMAP algorithm is a highly successful clustering tool that involves interesting ideas from mathematics and data science:

1) Spivak's theory of extended pseudo metric spaces (ep-metric spaces) 2) TDA constructions in ep-metric spaces 3) weighted graphs 4) classical dimension reduction 5) graph optimization: fuzzy sets, cross entropy

I will explain the algorithm from a mathematical point of view.

algebraic topologycategory theorygroup theoryK-theory and homology

Audience: researchers in the topic

Cihan Okay

Series comments: Contact the organizer to get access to Zoom.

Recordings of talks available at www.youtube.com/channel/UCLrmyGpqxyeVpTcA1b5HcMw/videos