ProximalGradient (tetrad-lib 7.6.7-SNAPSHOT API)

java.lang.Object

edu.pitt.csb.mgm.ProximalGradient

public class ProximalGradient extends Object

Implementation of Nesterov's 83 method as described in Beck and Teboulle, 2009 aka Fast Iterative Shrinkage Thresholding Algorithm

with step size scaling from Becker et all 2011

Created by ajsedgewick on 7/29/15.

Version:: $Id: $Id
Author:: josephramsey

Constructor Summary

Constructors

Constructor

Description

ProximalGradient()

Constructor using defaults from Becker et al 2011.

ProximalGradient(double beta, double alpha, boolean edgeConverge)

Constructor, set parameters for a proximal gradient run
Method Summary

Modifier and Type

Method

Description

cern.colt.matrix.DoubleMatrix1D

learnBackTrack(ConvexProximal cp, cern.colt.matrix.DoubleMatrix1D Xin, double epsilon, int iterLimit)

learnBackTrack.

static double

norm2(cern.colt.matrix.DoubleMatrix1D vec)

norm2.

void

setEdgeChangeTol(int t)

Positive edge change tolerance is the number of iterations with 0 edge changes needed to converge.

Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details
- ProximalGradient
  
  public ProximalGradient(double beta, double alpha, boolean edgeConverge)
  
  Constructor, set parameters for a proximal gradient run
  
  Parameters:
  
  beta - (0,1) factor to increase L when Lipshitz violated, L = L_old/beta
  
  alpha - (0,1) factor to decrease L otherwise, L = L_old*alpha
  
  edgeConverge - a boolean
- ProximalGradient
  
  public ProximalGradient()
  
  Constructor using defaults from Becker et al 2011. beta = .5, alpha = .9
Method Details
- norm2
  
  public static double norm2(cern.colt.matrix.DoubleMatrix1D vec)
  
  norm2.
  
  Parameters:
  
  vec - a DoubleMatrix1D object
  
  Returns:
  
  a double
- setEdgeChangeTol
  
  public void setEdgeChangeTol(int t)
  
  Positive edge change tolerance is the number of iterations with 0 edge changes needed to converge. Negative edge change tolerance means convergence happens when number of difference edges <= |edge change tol|. Default is 3.
  
  Parameters:
  
  t - a int
- learnBackTrack
  
  public cern.colt.matrix.DoubleMatrix1D learnBackTrack(ConvexProximal cp, cern.colt.matrix.DoubleMatrix1D Xin, double epsilon, int iterLimit)
  
  learnBackTrack.
  
  Parameters:
  
  cp - a ConvexProximal object
  
  Xin - a DoubleMatrix1D object
  
  epsilon - a double
  
  iterLimit - a int
  
  Returns:
  
  a DoubleMatrix1D object