Computational Linear Algebra 2: Topic Modelling with SVD & NMF

Course materials available here:
We use a dataset of messages posted on discussion forums to identify topics. A term-document matrix represents the frequency of the vocabulary in the documents. We factor it using Singular Value Decomposition (SVD) and Non-negative Matrix Factorization (NMF).

We use PyTorch as a GPU-accelerated alternative to Numpy to speed things up, and we cover Stochastic Gradient Descent, a very useful, general purpose optimization algorithm.

This video is fast-paced, so be sure to watch Lesson 3 for a review and Q&A of the topics covered here.

Course overview blog post:
Taught in the University of San Francisco MS in Analytics (MSAN) graduate program:
Ask questions about the course on our forums:

Topics covered:
– Singular Value Decomposition (SVD)
– Non-negative Matrix Factorization (NMF)
– Stochastic Gradient Descent (SGD)
– Intro to PyTorch


%d bloggers like this: