BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Paul Samuel Ignacio (University of the Philippines Baguio)
DTSTART:20200925T050000Z
DTEND:20200925T060000Z
DTSTAMP:20260423T004037Z
UID:APATG/4
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/APATG/4/">An
 other Topological "Reading" Lesson: Classification of MNIST using Bottlene
 ck-based Statistical Features</a>\nby Paul Samuel Ignacio (University of t
 he Philippines Baguio) as part of Asia Pacific Seminar on Applied Topology
  and Geometry\n\n\nAbstract\nInspired by the work of Garin and Tauzin on t
 he classification of grayscale images using features from a wide array of 
 topological summaries\, we perform classification of the MNIST data set us
 ing only features derived from statistics on bottleneck distances. While t
 here exist several critiques on the bottleneck metric\, we show that it ca
 n be used to produce features on which machine learning algorithms\, in th
 is case a random forest\, can be trained to produce respectable accuracy.\
 n
LOCATION:https://researchseminars.org/talk/APATG/4/
END:VEVENT
END:VCALENDAR