Reinforcement Learning Toolbox 2.0
last updated:
General
Documentation
Manual
Tutorial
Class Reference
Master Thesis
Examples
Related Papers
Downloads
Links
News
mailto:webmaster
Main Page     Class Hierarchy   Compound List   File List   Compound Members   File Members

CSemiMDPRewardListener Class Reference

Represents SMDP Listener which also need a reward. More...

#include <cagentlistener.h>

Inheritance diagram for CSemiMDPRewardListener:

CSemiMDPListener CParameterObject CParameters CAdaptiveParameterFromAverageRewardCalculator CAverageReinforcementBaseLineCalculator CEpisodeMatlabOutput CEpisodeOutput CFeatureRewardModel CFeatureStateRewardModel CGPOMDPGradientCalculator CGraphDebugger CLSTDLambda CPolicyEvaluator CREINFORCELearner CRewardEpisode CRewardLogger CSamplingBasedTransitionModel CTDLearner CVFunctionLearner CVPolicyLearner CVPolicyLearner List of all members.


Public Member Functions

  CSemiMDPRewardListener (CRewardFunction *semiMDPRewardFunction)
virtual void  nextStep (CStateCollection *oldState, CAction *action, CStateCollection *nextState)
  Calculates the reward and then calls nextStep(...) with the reward as additional argument.

virtual void  nextStep (CStateCollection *, CAction *, double, CStateCollection *)
  virtual function, to be implemented by subclass

virtual void  intermediateStep (CStateCollection *oldState, CAction *action, CStateCollection *nextState)
  Calculates the reward and then calls intermediateStep(...) with the reward as additional argument.

virtual void  intermediateStep (CStateCollection *, CAction *, double, CStateCollection *)
  virtual function, to be implemented by subclass

void  setRewardFunction (CRewardFunction *semiMDPRewardFunction)
CRewardFunction getRewardFunction ()


Protected Attributes

CRewardFunction semiMDPRewardFunction
  reward function for reward calculation


Detailed Description

Represents SMDP Listener which also need a reward.

The CSemiMDPRewardListener maintains a reward function. With this reward function, each time a nextStep or an intermediateStep event occurs the listener can calculate the reward and then he calls the specific abstract event function with the S-A-R(eward)-S tuple.


Constructor & Destructor Documentation

CSemiMDPRewardListener::CSemiMDPRewardListener CRewardFunction semiMDPRewardFunction  ) 
 
Parameters:
semiMDPRewardFunction  reward function for reward calculation

Member Function Documentation

CRewardFunction* CSemiMDPRewardListener::getRewardFunction  ) 
 
virtual void CSemiMDPRewardListener::intermediateStep CStateCollection ,
CAction ,
double  ,
CStateCollection
[inline, virtual]
 

virtual function, to be implemented by subclass

Reimplemented in CEpisodeOutput, CTDLearner, and CVFunctionLearner.

virtual void CSemiMDPRewardListener::intermediateStep CStateCollection oldState,
CAction action,
CStateCollection nextState
[virtual]
 

Calculates the reward and then calls intermediateStep(...) with the reward as additional argument.

Reimplemented from CSemiMDPListener.

virtual void CSemiMDPRewardListener::nextStep CStateCollection ,
CAction ,
double  ,
CStateCollection
[inline, virtual]
 

virtual function, to be implemented by subclass

Reimplemented in CAdaptiveParameterFromAverageRewardCalculator, CEpisodeOutput, CEpisodeMatlabOutput, CEpisodeOutputStateChanged, CPolicyEvaluator, CAverageRewardCalculator, CRewardPerEpisodeCalculator, CValueCalculator, CPolicySameStateEvaluator, CAverageRewardSameStateCalculator, CValueSameStateCalculator, CPolicyGreedynessEvaluator, CLSTDLambda, CGPOMDPGradientCalculator, CAverageReinforcementBaseLineCalculator, CREINFORCELearner, CFeatureRewardModel, CFeatureStateRewardModel, CRewardEpisode, CRewardLogger, CSamplingBasedTransitionModel, CSamplingBasedTransitionModelFromTransitionFunction, CGraphDebugger, CTDLearner, CVPolicyLearner, CVPolicyLearner, and CVFunctionLearner.

virtual void CSemiMDPRewardListener::nextStep CStateCollection oldState,
CAction action,
CStateCollection nextState
[virtual]
 

Calculates the reward and then calls nextStep(...) with the reward as additional argument.

Reimplemented from CSemiMDPListener.

Reimplemented in CSamplingBasedTransitionModel.

void CSemiMDPRewardListener::setRewardFunction CRewardFunction semiMDPRewardFunction  ) 
 

Member Data Documentation

CRewardFunction* CSemiMDPRewardListener::semiMDPRewardFunction [protected]
 

reward function for reward calculation


The documentation for this class was generated from the following file: