Money: 15~25 kSEK
(placeholer, TBC)
TITLE 'Non-negative Matrix Factorisation: theory and applications.'
LECTURER: Xiao Fu (Electrical Engineering and Computer Science, Oregon
State University, US).
TOPIC: Non-Negative matrix factorisation aims to decompose a data
matrix into low-rank latent factor matrices with non-negativity
constraints. This series of talks will focus on modern research on
its identifiability guarantees: What are the conditions that this is
possible? What are the conditions under which a certain algorithm
recovers the latent factors effectively? The approaches are
illustrated with examples drawn from the area of signal processing,
tensor analysis and machine learning.
TIMING: 5 tutorial seminars, Spring 2018.