OpenANN::ActionSpace Class Reference

Represents the action space in a reinforcement learning problem. More...

#include <ActionSpace.h>

Inheritance diagram for OpenANN::ActionSpace:

Public Types
typedef Eigen::VectorXd	Action
typedef std::vector< Action >	A

Public Member Functions
virtual	~ActionSpace ()
virtual int	actionSpaceDimension () const =0
virtual bool	actionSpaceContinuous () const =0
virtual int	actionSpaceElements () const =0
virtual const Action &	actionSpaceLowerBound () const =0
virtual const Action &	actionSpaceUpperBound () const =0
virtual const A &	getDiscreteActionSpace () const =0

Detailed Description

Represents the action space in a reinforcement learning problem.

The action space contains all possible actions that the agent can take.

Member Typedef Documentation

typedef std::vector<Action> OpenANN::ActionSpace::A

typedef Eigen::VectorXd OpenANN::ActionSpace::Action

Constructor & Destructor Documentation

virtual OpenANN::ActionSpace::~ActionSpace ( )

inlinevirtual

Member Function Documentation

virtual bool OpenANN::ActionSpace::actionSpaceContinuous ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

virtual int OpenANN::ActionSpace::actionSpaceDimension ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

virtual int OpenANN::ActionSpace::actionSpaceElements ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

virtual const Action& OpenANN::ActionSpace::actionSpaceLowerBound ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

virtual const Action& OpenANN::ActionSpace::actionSpaceUpperBound ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

virtual const A& OpenANN::ActionSpace::getDiscreteActionSpace ( ) const

pure virtual

Implemented in DoublePoleBalancing, and SinglePoleBalancing.

---

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

• OpenANN/rl/ActionSpace.h