scalation/scalation/scalation.optimization/LassoAdmm

LassoAdmm

scalation.optimization.LassoAdmm

The LassoAdmm object performs LASSO regression using Alternating Direction Method of Multipliers (ADMM). Minimize the following objective function to find an optimal solutions for x.

argmin_x (1/2)||Ax − b||_2^2 + λ||x||_1

A = data matrix
b = response vector
λ = weighting on the l_1 penalty
x = solution (coefficient vector)

Attributes

See also: euler.stat.yale.edu/~tba3/stat612/lectures/lec23/lecture23.pdf

https://web.stanford.edu/~boyd/papers/admm_distr_stats.html
Graph
Supertypes: class Object

trait Matchable

class Any
Self type: LassoAdmm.type

Members list

Value members

Concrete methods

Reset the warm start map.

Attributes

Solve for x using ADMM.

Value parameters

a: the data matrix
b: the response vector
λ: the regularization l_1 penalty weight

Attributes

Solve for x using ADMM using cached factorizations for efficiency.

Value parameters

ata_ρI_inv: cached (a.t * a + ρI)^-1
atb: cached a.t * b
λ: the regularization l_1 penalty weight

Attributes

Concrete fields

In this article

Generated with