burlap.behavior.singleagent.planning.vfa.fittedvi

Interface SupervisedVFA

All Known Implementing Classes:

WekaVFATrainer

public interface SupervisedVFA

An interface for learning value function approximation via a supervised learning algorithm. This interface defines the method train(java.util.List<burlap.behavior.singleagent.planning.vfa.fittedvi.SupervisedVFA.SupervisedVFAInstance>) which takes as input a list of SupervisedVFA.SupervisedVFAInstance objects, runs a regression algorithm, and returns the learned function, which is an interface of ValueFunction.

A SupervisedVFA.SupervisedVFAInstance is a pair consisting of a State and the target state value that is to be learned.

Author:

James MacGlashan.

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	SupervisedVFA.SupervisedVFAInstance A pair for a state and it's target value function value.

Method Summary

Methods

Modifier and Type	Method and Description
ValueFunction	train(java.util.List<SupervisedVFA.SupervisedVFAInstance> trainingData) Uses supervised learning (regression) to learn a value function approximation of the input training data.

Method Detail

train

ValueFunction train(java.util.List<SupervisedVFA.SupervisedVFAInstance> trainingData)

Uses supervised learning (regression) to learn a value function approximation of the input training data.

Parameters:

trainingData - the training data to fit.

Returns:

a ValueFunction that fits the training data.