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 CRewardFunction Class Reference Class for clalculating the reward for the learning objects. More... #include <crewardfunction.h> Inheritance diagram for CRewardFunction: List of all members. Public Member Functions virtual  ~CRewardFunction () virtual double  getReward (CStateCollection *oldState, CAction *action, CStateCollection *newState)=0   Virtual function for calculating the reward. Detailed Description Class for clalculating the reward for the learning objects. The reward is calculated for each learning object separately. The reward function objects have to return a double-valued reward given an S-A-S tuple, this is done by the function getReward. So for an own learning problem this class has to be derivated to implement the getReward function. If you need access to the action index, you have to provide an actionset for your reward function. Each listener who needs a reward is subclass of CSemiMDPRewardListener, which maintains a reward function object and calculates the reward for the listner. See also: CSemiMDPRewardListener Constructor & Destructor Documentation virtual CRewardFunction::~CRewardFunction (   )  [inline, virtual] Member Function Documentation virtual double CRewardFunction::getReward (  CStateCollection *  oldState, CAction *  action, CStateCollection *  newState )  [pure virtual]   Virtual function for calculating the reward. Implemented in CFunctionComperator, CGridWorldModel, CMultiPoleModel, CFeatureRewardFunction, CStateReward, CZeroReward, CRewardFunctionFromValueFunction, CSemiMDPLastNRewardFunction, CTaxiDomain, CTaxiHierarchicalBehaviour, and CExtendedActionTransitionFunction. The documentation for this class was generated from the following file: crewardfunction.h