One-dimensional total variation denoising
= One-dimensional total variation denoising
{title2=$\widehat\mu_\lambda=\arg\min_\mu\{\|Y-\mu\|_2^2/(2n)+\lambda\sum_i|\mu_{i+1}-\mu_i|\}$}
= One-dimensional fused Lasso
{synonym}
This version of the <fused Lasso> penalizes only adjacent differences. It estimates a piecewise-constant signal while preserving an unpenalized constant level. Larger penalties generally favour fewer fitted changes.