Group Theory and Machine Learning
The use of algebraic methods—specifically group theory, representation theory, and even some concepts from algebraic geometry—is an emerging new direction in machine learning. The purpose of this tutorial is to give an entertaining but informative introduction to the background to these developments and sketch some of the many possible applications, including multi-object tracking, learning rankings, and constructing translation and rotation invariant features for image recognition. The tutorial is intended to be palatable by a non-specialist audience with no prior background in abstract algebra.
Be forewarned, tough sledding if you are not already a machine learning sort of person.
But, since I don’t post what I haven’t watched, I did watch the entire video.
It suddenly got interesting just past 93:08 when Risi Kondor started talking about blobs on radar screens and associating information with them…., wait, run that by once again, …blobs on radar screens and associating information with them.
Oh, that is what I thought he said.
I suppose for fire control systems and the like as well as civilian applications.
I am so much of a text and information navigation person that I don’t often think about other applications for “pattern recognition” and the like.
With all the international traveling I used to do, being a blob on a radar screen got my attention!
Has applications in tracking animals in the wild and other tracking with sensor data.
Another illustration of why topic maps need an open-ended and extensible notion of subject identification.
What we think of as methods of subject identification may not be what others think of as methods of subject identification.