Between natural metric and natural-vector-fitted manifold
Organizer
Speaker
(Tsinghua) Tao Zhou
Time
Thursday, October 26, 2023 8:00 PM - 8:30 PM
Venue
Online
Abstract
The manifold hypothesis posits that high-dimensional data typically lie close to a low-dimensional manifold. By applying the streamer fitting technique to natural vectors, we can obtain the geodetic distance on the fitted manifold. We are interested in the relationship between this geodetic distance and the natural metric: do they behave similarly on specific datasets (although they necessarily differ in magnitude)? How will conducting phylogenetic analyses using these two distances differ? I will explain the above questions with some example analyses.