de.jungblut.online.regularization

Class GradientDescentUpdater

java.lang.Object

de.jungblut.online.regularization.GradientDescentUpdater

All Implemented Interfaces:

WeightUpdater

Direct Known Subclasses:

AdaptiveFTRLRegularizer, L1Regularizer, L2Regularizer

public class GradientDescentUpdater
extends Object
implements WeightUpdater

Constructor Summary

Constructors

Constructor and Description
GradientDescentUpdater()

Method Summary

Modifier and Type	Method and Description
de.jungblut.math.minimize.CostGradientTuple	computeGradient(de.jungblut.math.DoubleVector theta, de.jungblut.math.DoubleVector gradient, double learningRate, long iteration, double lambda, double cost) Computes the gradient.
CostWeightTuple	computeNewWeights(de.jungblut.math.DoubleVector theta, de.jungblut.math.DoubleVector gradient, double learningRate, long iteration, double lambda, double cost) Simplistic gradient descent without regularization.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

GradientDescentUpdater

public GradientDescentUpdater()

Method Detail

computeNewWeights

public CostWeightTuple computeNewWeights(de.jungblut.math.DoubleVector theta,
                                         de.jungblut.math.DoubleVector gradient,
                                         double learningRate,
                                         long iteration,
                                         double lambda,
                                         double cost)

Simplistic gradient descent without regularization.

Specified by:

computeNewWeights in interface WeightUpdater

Parameters:

theta - the old weights.

gradient - the gradient.

learningRate - the learning rate.

iteration - the number of the current iteration.

lambda - the regularization parameter.

cost - the computed cost for this gradient update.

Returns:

the already updated weights for a particular updated gradient.

computeGradient

public de.jungblut.math.minimize.CostGradientTuple computeGradient(de.jungblut.math.DoubleVector theta,
                                                                   de.jungblut.math.DoubleVector gradient,
                                                                   double learningRate,
                                                                   long iteration,
                                                                   double lambda,
                                                                   double cost)

Description copied from interface: WeightUpdater

Computes the gradient.

Specified by:

computeGradient in interface WeightUpdater

Parameters:

theta - the old weights.

gradient - the gradient.

learningRate - the learning rate.

iteration - the number of the current iteration.

lambda - the regularization parameter.

cost - the computed cost for this gradient update.

Returns:

the gradient vector that should be substracted from the weights and the updated cost.