burlap.behavior.policy

Class EpsilonGreedy

java.lang.Object

burlap.behavior.policy.EpsilonGreedy

All Implemented Interfaces:

EnumerablePolicy, Policy, SolverDerivedPolicy

public class EpsilonGreedy
extends java.lang.Object
implements SolverDerivedPolicy, EnumerablePolicy

This class defines a an epsilon-greedy policy over Q-values and requires a QComputable valueFunction to be specified. With probability epsilon the policy will return a random action (with uniform distribution over all possible action). With probability 1 - epsilon the policy will return the greedy action. If multiple actions tie for the highest Q-value, then one of the tied actions is randomly selected.

Author:

James MacGlashan

Field Summary

Fields

Modifier and Type	Field and Description
protected double	epsilon
protected QProvider	qplanner
protected java.util.Random	rand

Constructor Summary

Constructors

Constructor and Description
EpsilonGreedy(double epsilon) Initializes with the value of epsilon, where epsilon is the probability of taking a random action.
EpsilonGreedy(QProvider planner, double epsilon) Initializes with the QComputablePlanner to use and the value of epsilon to use, where epsilon is the probability of taking a random action.

Method Summary

Modifier and Type	Method and Description
Action	action(State s) This method will return an action sampled by the policy for the given state.
double	actionProb(State s, Action a) Returns the probability/probability density that the given action will be taken in the given state.
boolean	definedFor(State s) Specifies whether this policy is defined for the input state.
double	getEpsilon() Returns the epsilon value, where epsilon is the probability of taking a random action.
java.util.List<ActionProb>	policyDistribution(State s) This method will return action probability distribution defined by the policy.
void	setEpsilon(double epsilon) Sets the epsilon value, where epsilon is the probability of taking a random action.
void	setSolver(MDPSolverInterface solver) Sets the valueFunction whose results affect this policy.

Methods inherited from class java.lang.Object

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

Field Detail

qplanner

protected QProvider qplanner

epsilon

protected double epsilon

rand

protected java.util.Random rand

Constructor Detail

EpsilonGreedy

public EpsilonGreedy(double epsilon)

Initializes with the value of epsilon, where epsilon is the probability of taking a random action.

Parameters:

epsilon - the probability of taking a random action.

EpsilonGreedy

public EpsilonGreedy(QProvider planner,
                     double epsilon)

Initializes with the QComputablePlanner to use and the value of epsilon to use, where epsilon is the probability of taking a random action.

Parameters:

planner - the QComputablePlanner to use

epsilon - the probability of taking a random action.

Method Detail

getEpsilon

public double getEpsilon()

Returns the epsilon value, where epsilon is the probability of taking a random action.

Returns:

the epsilon value

setEpsilon

public void setEpsilon(double epsilon)

Sets the epsilon value, where epsilon is the probability of taking a random action.

Parameters:

epsilon - the probability of taking a random action.

setSolver

public void setSolver(MDPSolverInterface solver)

Description copied from interface: SolverDerivedPolicy

Sets the valueFunction whose results affect this policy.

Specified by:

setSolver in interface SolverDerivedPolicy

Parameters:

solver - the solver from which this policy is derived

action

public Action action(State s)

Description copied from interface: Policy

This method will return an action sampled by the policy for the given state. If the defined policy is stochastic, then multiple calls to this method for the same state may return different actions. The sampling should be with respect to defined action distribution that is returned by getActionDistributionForState

Specified by:

action in interface Policy

Parameters:

s - the state for which an action should be returned

Returns:

a sample action from the action distribution; null if the policy is undefined for s

actionProb

public double actionProb(State s,
                         Action a)

Description copied from interface: Policy

Returns the probability/probability density that the given action will be taken in the given state.

Specified by:

actionProb in interface Policy

Parameters:

s - the state of interest

a - the action that may be taken in the state

Returns:

the probability/probability density

policyDistribution

public java.util.List<ActionProb> policyDistribution(State s)

Description copied from interface: EnumerablePolicy

This method will return action probability distribution defined by the policy. The action distribution is represented by a list of ActionProb objects, each which specifies a grounded action and a probability of that grounded action being taken. The returned list does not have to include actions with probability 0.

Specified by:

policyDistribution in interface EnumerablePolicy

Parameters:

s - the state for which an action distribution should be returned

Returns:

a list of possible actions taken by the policy and their probability.

definedFor

public boolean definedFor(State s)

Description copied from interface: Policy

Specifies whether this policy is defined for the input state.

Specified by:

definedFor in interface Policy

Parameters:

s - the input state to test for whether this policy is defined

Returns:

true if this policy is defined for State s, false otherwise.