burlap.behavior.singleagent.learnfromdemo.mlirl.differentiableplanners

Class DifferentiableSparseSampling.DiffStateNode

java.lang.Object

burlap.behavior.singleagent.learnfromdemo.mlirl.differentiableplanners.DifferentiableSparseSampling.DiffStateNode

Enclosing class:

DifferentiableSparseSampling

public class DifferentiableSparseSampling.DiffStateNode
extends java.lang.Object

A class for value differentiable state nodes. Includes the state, a value estimate, whether the node has been closed and methods for estimating the Q and V values.

Author:

James MacGlashan

Constructor Summary

Constructors

Constructor and Description
DiffStateNode(HashableState sh, int height)

Method Summary

Modifier and Type	Method and Description
DifferentiableSparseSampling.QAndQGradient	estimateQs()
DifferentiableSparseSampling.VAndVGradient	estimateV()
void	exactQEstimate(Action ga, DifferentiableSparseSampling.QAndQGradient qs)
void	sampledQEstimate(Action ga, DifferentiableSparseSampling.QAndQGradient qs)
protected void	setV(DifferentiableSparseSampling.QAndQGradient qvs)
protected void	setVGrad(DifferentiableSparseSampling.QAndQGradient qvs)

Methods inherited from class java.lang.Object

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

Constructor Detail

DiffStateNode

public DiffStateNode(HashableState sh,
                     int height)

Method Detail

estimateQs

public DifferentiableSparseSampling.QAndQGradient estimateQs()

sampledQEstimate

public void sampledQEstimate(Action ga,
                             DifferentiableSparseSampling.QAndQGradient qs)

exactQEstimate

public void exactQEstimate(Action ga,
                           DifferentiableSparseSampling.QAndQGradient qs)

estimateV

public DifferentiableSparseSampling.VAndVGradient estimateV()

setV

protected void setV(DifferentiableSparseSampling.QAndQGradient qvs)

setVGrad

protected void setVGrad(DifferentiableSparseSampling.QAndQGradient qvs)