News for hackers. Stuff that matters.
Re:Separability, SVD and low-rank approximation of 2D image ... by jacobolus on Monday March 02, 2020 @01:45AM (#22462224)
Here’s a relevant recent lecture from Alex Townsend, as part of Gil Strang’s 18.065 course, “Rapidly Decreasing Singular Values” https://www.youtube.com/watch?v=9BYsNpTCZGg
Re:Separability, SVD and low-rank approximation of 2D image ... by dispanser on Monday March 02, 2020 @10:20AM (#22465077)
Currently watching the lecture series, and it's a great experience. In fact, this made me understand the blog post to some level I wouldn't have thought possible only a few weeks ago.
Re:Separability, SVD and low-rank approximation of 2D image ... by kiriliponi on Monday March 02, 2020 @08:57AM (#22464421)
Very enjoyable, thanks for sharing
Re:Separability, SVD and low-rank approximation of 2D image ... by noelwelsh on Monday March 02, 2020 @04:30AM (#22462941)
Good stuff. A lot of clustering methods can be viewed as matrix factorization (see, e.g., https://arxiv.org/abs/1512.07548) so it would be interesting to see if you could spend more computation on the creating the low dimensional approximation to achieve a better result. E.g. perhaps optimizing the absolute error (L1 norm) would be better than the squared error (L2 norm)?
Re:Separability, SVD and low-rank approximation of 2D image ... by id_ris on Monday March 02, 2020 @12:01PM (#22466213)
Very nice writeup. It's fun to see linear algebra out in the wild.