As many of you know, the Fused Lasso is one of well known penalized methods, which is introduced by Tibshirani, 2005.
However, I don't get to the meaning of how it is called.
Could anyone give any ideas about the origin or the meaning of 'fused'?
fused-lassolassomany-categoriesregularizationterminology
As many of you know, the Fused Lasso is one of well known penalized methods, which is introduced by Tibshirani, 2005.
However, I don't get to the meaning of how it is called.
Could anyone give any ideas about the origin or the meaning of 'fused'?
Best Answer
The term is clearly explained in the abstract of the paper[1] you mention.
That is, it's called that because adjacent parameters may be set equal -- i.e. "fused" (somewhat akin to aligning broken bones, which ultimately fuse together).
This is the first meaning of fuse here "to join or blend to form a single entity".
Diagrams in the paper further highlight the "fusion" of adjacent parameters.
[1] Tibshirani et al (2005),
"Sparsity and smoothness via the fused lasso,"
Journal of the Royal Statistical Society, Series B, 67, 91–108