burlap.behavior.singleagent

Interface ValueFunctionInitialization

All Superinterfaces:

ValueFunction

All Known Subinterfaces:

DifferentiableVInit

All Known Implementing Classes:

DifferentiableVInit.ParamedDiffVInit, FittedVI.VFAVInit, LinearDiffRFVInit, LinearStateDiffVF, ValueFunctionInitialization.ConstantValueFunctionInitialization, VanillaDiffVinit

public interface ValueFunctionInitialization
extends ValueFunction

This interface may be used by planning and learning algorithms that require an initialization value for the Q-value function or the value function. A common implementation for initializing all values to the same constant is provided. This class extends the ValueFunction class, so the initialization for the value function. may be retrieved with the standard ValueFunction.value(burlap.oomdp.core.State) method. It also adds a qValue(burlap.oomdp.core.State, burlap.oomdp.core.AbstractGroundedAction) method for initializing Q-values.

Author:

James MacGlashan

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	ValueFunctionInitialization.ConstantValueFunctionInitialization A ValueFunctionInitialization implementation that always returns a constant value.

Method Summary

Methods

Modifier and Type	Method and Description
double	qValue(State s, AbstractGroundedAction a) Returns the initialization value of the Q-value function for a given state and action pair.

Methods inherited from interface burlap.behavior.singleagent.planning.ValueFunction

value

Method Detail

qValue

double qValue(State s,
            AbstractGroundedAction a)

Returns the initialization value of the Q-value function for a given state and action pair.

Parameters:

s - the state for which to get the initial value of the Q-value function.

a - the action for which to get the initial value of the Q-value function.

Returns:

the initialization value of the Q-value function for a given state and action pair.