mlpack::rl::RewardClipping< EnvironmentType > Class Template Reference

Interface for clipping the reward to some value between the specified maximum and minimum value (Clipping here is implemented as \( g_{\text{clipped}} = \max(g_{\text{min}}, \min(g_{\text{min}}, g))) \).) More...

#include <reward_clipping.hpp>

Public Types
using	State = typename EnvironmentType::State
	Convenient typedef for state.
using	Action = typename EnvironmentType::Action
	Convenient typedef for action.

Public Member Functions
	RewardClipping (EnvironmentType &environment, const double minReward=-1.0, const double maxReward=1.0)
	Constructor for creating a RewardClipping instance. More...
State	InitialSample ()
	The InitialSample method is called by the environment to initialize the starting state. More...
bool	IsTerminal (const State &state) const
	Checks whether given state is a terminal state. More...
double	Sample (const State &state, const Action &action, State &nextState)
	Dynamics of Environment. More...
double	Sample (const State &state, const Action &action)
	Dynamics of Environment. More...
EnvironmentType &	Environment () const
	Get the environment.
EnvironmentType &	Environment ()
	Modify the environment.
double	MinReward () const
	Get the minimum reward value.
double &	MinReward ()
	Modify the minimum reward value.
double	MaxReward () const
	Get the maximum reward value.
double &	MaxReward ()
	Modify the maximum reward value.

Detailed Description

template<typename EnvironmentType>
class mlpack::rl::RewardClipping< EnvironmentType >

Interface for clipping the reward to some value between the specified maximum and minimum value (Clipping here is implemented as \( g_{\text{clipped}} = \max(g_{\text{min}}, \min(g_{\text{min}}, g))) \).)

Template Parameters

EnvironmentType

A type of Environment that is being wrapped.

Constructor & Destructor Documentation

RewardClipping()

template<typename EnvironmentType >

mlpack::rl::RewardClipping< EnvironmentType >::RewardClipping ( EnvironmentType &  environment, const double  minReward = -1.0, const double  maxReward = 1.0  )

inline

Constructor for creating a RewardClipping instance.

Parameters

minReward	Minimum possible value of clipped reward.
maxReward	Maximum possible value of clipped reward.
environment	An instance of the environment used for actual simulations.

Member Function Documentation

InitialSample()

template<typename EnvironmentType >

State mlpack::rl::RewardClipping< EnvironmentType >::InitialSample ( )

inline

The InitialSample method is called by the environment to initialize the starting state.

Returns whatever Initial Sample is returned by the environment.

IsTerminal()

template<typename EnvironmentType >

bool mlpack::rl::RewardClipping< EnvironmentType >::IsTerminal ( const State &  state ) const

inline

Checks whether given state is a terminal state.

Returns the value by calling the environment method.

Parameters

state

desired state.

Returns

true if state is a terminal state, otherwise false.

Sample() [1/2]

template<typename EnvironmentType >

double mlpack::rl::RewardClipping< EnvironmentType >::Sample ( const State &  state, const Action &  action, State &  nextState  )

inline

Dynamics of Environment.

The rewards returned from the base environment are clipped according the maximum and minimum values specified.

Parameters

state	The current state.
action	The current action.
nextState	The next state.

Returns

clippedReward, Reward clipped between [minReward, maxReward].

Sample() [2/2]

template<typename EnvironmentType >

double mlpack::rl::RewardClipping< EnvironmentType >::Sample ( const State &  state, const Action &  action  )

inline

Dynamics of Environment.

The rewards returned from the base environment are clipped according the maximum and minimum values specified.

Parameters

state	The current state.
action	The current action.

Returns

clippedReward, Reward clipped between [minReward, maxReward].

---

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

• src/mlpack/methods/reinforcement_learning/environment/reward_clipping.hpp