Computational Linear Algebra 2: Topic Modelling with SVD & NMF

Course materials available here: https://github.com/fastai/numerical-linear-algebra
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: http://www.fast.ai/2017/07/17/num-lin-alg/
Taught in the University of San Francisco MS in Analytics (MSAN) graduate program: https://www.usfca.edu/arts-sciences/graduate-programs/analytics
Ask questions about the course on our fast.ai forums: http://forums.fast.ai/c/lin-alg

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

Questions?

%d bloggers like this: