LatentVariableRegression: Fit a latent variable linar multi-task model (Zhang et al....

Description Usage Arguments Value See Also

Description

A Bayesian linear multi-task model where the regression matrix is assumed to be composed of latent factors.

Usage

1
2
3
LatentVariableRegression(X = NULL, task.specific.features = list(), Y, S,
  max.iter = 10000, epsilon = 1e-05, XTX = NULL, XTY = NULL,
  verbose = 1)

Arguments

X

N by J1 matrix of features common to all tasks.

task.specific.features

List of features which are specific to each task. Each entry contains an N by J2 matrix for one particular task (where columns are features). List has to be ordered according to the columns of Y.

Y

N by K output matrix for every task.

S

H by K loading matrix.

max.iter

(Optional) Maximum number of iterations.

epsilon

(Optional) Desired accuracy. If error change drops below epsilon, the algorithm terminates.

XTX

(Optional) Precomputed matrices t(X)*X as for example produced by PrepareMatrices.

XTY

(Optional) Precomputed matrices t(X)*Y as for example produced by PrepareMatrices

verbose

(Optional) Integer in 0,1,2. verbose = 0: No output. verbose = 1: Print summary at the end of the optimization. verbose = 2: Print progress during optimization.

Value

List containing

Gamma

Estimated mixing matrix.

sigma2

Estimated sigma^2.

Psi

Estimated Psi.

S

Loading matrix used.

B

MAP estimate of the regression coefficients.

early.termination

Boolean indicating whether the algorithm exceeded max.iter iterations.

See Also

RunGroupCrossvalidation


tohein/linearMTL documentation built on May 17, 2019, 8:22 a.m.