burlap.behavior.singleagent.learnfromdemo

Class IRLRequest

java.lang.Object

burlap.behavior.singleagent.learnfromdemo.IRLRequest

Direct Known Subclasses:

ApprenticeshipLearningRequest, MLIRLRequest

public class IRLRequest
extends java.lang.Object

A data structure for setting the common parameters necessary for an IRL algorithm.

Author:

Stephen Brawner and Mark Ho

Field Summary

Fields

Modifier and Type	Field and Description
protected SADomain	domain The domain in which IRL is to be performed
protected java.util.List<Episode>	expertEpisodes The input trajectories/episodes that are to be modeled.
protected double	gamma The discount factor of the problem
protected Planner	planner The planning algorithm used to compute the policy for a given reward function

Constructor Summary

Constructors

Constructor and Description
IRLRequest() Values will not be initialized.
IRLRequest(SADomain domain, Planner planner, java.util.List<Episode> expertEpisodes) Initializes.

Method Summary

Modifier and Type	Method and Description
SADomain	getDomain()
java.util.List<Episode>	getExpertEpisodes()
double	getGamma()
Planner	getPlanner()
boolean	isValid() Returns true if this request object has valid data members set; false otherwise.
void	setDomain(SADomain d)
void	setExpertEpisodes(java.util.List<Episode> episodeList)
void	setGamma(double gamma)
void	setPlanner(Planner p)

Methods inherited from class java.lang.Object

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

Field Detail

domain

protected SADomain domain

The domain in which IRL is to be performed

planner

protected Planner planner

The planning algorithm used to compute the policy for a given reward function

expertEpisodes

protected java.util.List<Episode> expertEpisodes

The input trajectories/episodes that are to be modeled.

gamma

protected double gamma

The discount factor of the problem

Constructor Detail

IRLRequest

public IRLRequest()

Values will not be initialized. You must use the setters for the domain, valueFunction, and expert episodes.

IRLRequest

public IRLRequest(SADomain domain,
                  Planner planner,
                  java.util.List<Episode> expertEpisodes)

Initializes. Discount factor will be defaulted to 0.99, which can optionally be changed with a setter.

Parameters:

domain - the domain in which IRL is to be performed

planner - the planning algorithm the IRL algorithm will invoke.

expertEpisodes - the example expert trajectories/episodes.

Method Detail

isValid

public boolean isValid()

Returns true if this request object has valid data members set; false otherwise.

Returns:

true if this request object has valid data members set; false otherwise.

setDomain

public void setDomain(SADomain d)

setPlanner

public void setPlanner(Planner p)

setExpertEpisodes

public void setExpertEpisodes(java.util.List<Episode> episodeList)

setGamma

public void setGamma(double gamma)

getDomain

public SADomain getDomain()

getPlanner

public Planner getPlanner()

getGamma

public double getGamma()

getExpertEpisodes

public java.util.List<Episode> getExpertEpisodes()