From e6c2b608f1b0cecbbba502ae103a4f4ce42fe3aa Mon Sep 17 00:00:00 2001 From: "Steven G. Johnson" Date: Mon, 17 Apr 2023 16:07:26 -0400 Subject: [PATCH] note elastic net --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 457160d..60ff16b 100644 --- a/README.md +++ b/README.md @@ -308,7 +308,7 @@ http://dx.doi.org/10.1137/S1052623499362822) — I used the "linear and separabl * Compressive sensing (CS) and ℓ¹ regularization (LASSO etc.). * [Slides](https://www.dropbox.com/s/iplx3gsk42t0xlb/Various-CS-slides.pdf?dl=0) collected from various sources. -**Further reading:** Strang textbook, section III.5. There are many tutorials and other information on CS/LASSO/etcetera online. For example, these [Rice Univ. tutorial slides (Cevher, 2019)](https://www.epfl.ch/labs/lions/wp-content/uploads/2019/01/Volkan-CS-IPSN09-tutorial-part-1.pdf) or [Princeton Slides (Cheng, 2014)](https://3dvision.princeton.edu/courses/COS598/2014sp/slides/lecture20_Compressive_Sensing.pdf) are fairly accessible. For compressed sensing in MRI, see e.g. the slides by [Lustig et al.](https://pages.cs.wisc.edu/~brecht/cs838docs/deshpande.project.pdf) and [Tamir (2019)](http://users.ece.utexas.edu/~jtamir/files/jtamir_compressed_sensing_ismrm19.pdf) and many other sources. +**Further reading:** Strang textbook, section III.5. There are many tutorials and other information on CS/LASSO/etcetera online. For example, these [Rice Univ. tutorial slides (Cevher, 2019)](https://www.epfl.ch/labs/lions/wp-content/uploads/2019/01/Volkan-CS-IPSN09-tutorial-part-1.pdf) or [Princeton Slides (Cheng, 2014)](https://3dvision.princeton.edu/courses/COS598/2014sp/slides/lecture20_Compressive_Sensing.pdf) are fairly accessible. For compressed sensing in MRI, see e.g. the slides by [Lustig et al.](https://pages.cs.wisc.edu/~brecht/cs838docs/deshpande.project.pdf) and [Tamir (2019)](http://users.ece.utexas.edu/~jtamir/files/jtamir_compressed_sensing_ismrm19.pdf) and many other sources. A hybrid of ℓ¹ (CS/LASSO) and ℓ² (ridge/Tikhonov) regularization is to use *both*, a combination called [elastic-net regularization](https://en.wikipedia.org/wiki/Elastic_net_regularization); see e.g. slides from [Univ. Iowa (Breheny)](https://myweb.uiowa.edu/pbreheny/7600/s16/notes/3-28.pdf). ## Lecture 26 (April 12)