TigerDomain

java.lang.Object
- burlap.domain.singleagent.pomdp.tiger.TigerDomain

All Implemented Interfaces:

DomainGenerator
```
public class TigerDomain
extends java.lang.Object
implements DomainGenerator
```
An implementation of the classic Tiger domain. In this problem an agent is faced with two closed doors side by side: a left door and a right door. Behind one is a prize (or something corresponding to high reward); behind the other is a tiger that will eat the agent. The goal is the for the agent to open the door with the prize. However, the catch is that the agent cannot know for certain which door the tiger is behind. To gain information, the agent can listen to the doors, which will produce a noisy observation regarding whether they hear the tiger behind the left or right door. Listening incurs a small reward cost, choosing the door with the prize gives a large positive reward, and choosing the door with the tiger incurs a large reward cost. This domain may optionally be specified to include a "do nothing" action which has no cost and provides no reward. It is useful when testing an algorithms willingness to take exploratory actions.

Field Summary

Fields
Modifier and Type	Field and Description
`static java.lang.String`	`ACTION_DO_NOTHING` The do nothing action name
`static java.lang.String`	`ACTION_LEFT` The open left door action name
`static java.lang.String`	`ACTION_LISTEN` The listen action name
`static java.lang.String`	`ACTION_RIGHT` The open right door action name
`double`	`correctDoorReward` the reward for opening the correct door
`static java.lang.String`	`DOOR_RESET` The observation value for when reaching a new pair of doors (occurs after opening a door)
`static java.lang.String`	`HEAR_LEFT` The observation attribute value for hearing the tiger behind the left door.
`static java.lang.String`	`HEAR_NOTHING` The observation of hearing nothing (occurs when taking the do nothing action)
`static java.lang.String`	`HEAR_RIGHT` The observation attribute value for hearing the tiger behind the right door
`protected boolean`	`includeDoNothing` Whether this domain should include the do nothing action or not
`protected double`	`listenAccuracy` The probability of hearing accurately where the tiger is
`double`	`listenReward` The reward for listening
`double`	`nothingReward` The reward for do nothing.
`static java.lang.String`	`VAL_LEFT` The discrete attribute value for the tiger being behind the left door
`static java.lang.String`	`VAL_RIGHT` The discrete attribtue value for the tiger being behind the right door
`static java.lang.String`	`VAR_DOOR` The attribute name that defines which door the tiger is behind
`static java.lang.String`	`VAR_HEAR` The variable key for an observation
`double`	`wrongDoorReward` The reward for opening the wrong door

Constructor Summary

Constructors
Constructor and Description
`TigerDomain()` Initializes.
`TigerDomain(boolean includeDoNothing)` Initializes.
`TigerDomain(boolean includeDoNothing, double listenAccuracy)` Initializes

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`Domain`	`generateDomain()` Returns a newly instanced Domain object
`double`	`getCorrectDoorReward()`
`static TabularBeliefState`	`getInitialBeliefState(PODomain domain)` Generates an initial `TabularBeliefState` in which the it is equally uncertain where the tiger is (50/50).
`double`	`getListenAccuracy()`
`double`	`getListenReward()`
`double`	`getNothingReward()`
`double`	`getWrongDoorReward()`
`boolean`	`isIncludeDoNothing()`
`static void`	`main(java.lang.String[] args)` Main method for interacting with the tiger domain via an `EnvironmentShell` By default, the TerminalExplorer interacts with the partially observable environment (`SimulatedPOEnvironment`), which means you only get to see the observations that the agent would.
`static StateGenerator`	`randomSideStateGenerator()` Returns a `StateGenerator` that 50% of the time generates an hidden tiger state with the tiger on the left side, and 50% time on the right.
`static StateGenerator`	`randomSideStateGenerator(double probLeft)` Returns a `StateGenerator` that some of the of the time generates an hidden tiger state with the tiger on the left side, and others on the right.
`void`	`setCorrectDoorReward(double correctDoorReward)`
`void`	`setIncludeDoNothing(boolean includeDoNothing)`
`void`	`setListenAccuracy(double listenAccuracy)`
`void`	`setListenReward(double listenReward)`
`void`	`setNothingReward(double nothingReward)`
`void`	`setWrongDoorReward(double wrongDoorReward)`

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - VAR_DOOR
```
public static final java.lang.String VAR_DOOR
```
    The attribute name that defines which door the tiger is behind
    
    See Also:
    
    Constant Field Values
  - VAR_HEAR
```
public static final java.lang.String VAR_HEAR
```
    The variable key for an observation
    
    See Also:
    
    Constant Field Values
  - ACTION_LEFT
```
public static final java.lang.String ACTION_LEFT
```
    The open left door action name
    
    See Also:
    
    Constant Field Values
  - ACTION_RIGHT
```
public static final java.lang.String ACTION_RIGHT
```
    The open right door action name
    
    See Also:
    
    Constant Field Values
  - ACTION_LISTEN
```
public static final java.lang.String ACTION_LISTEN
```
    The listen action name
    
    See Also:
    
    Constant Field Values
  - ACTION_DO_NOTHING
```
public static final java.lang.String ACTION_DO_NOTHING
```
    The do nothing action name
    
    See Also:
    
    Constant Field Values
  - VAL_LEFT
```
public static final java.lang.String VAL_LEFT
```
    The discrete attribute value for the tiger being behind the left door
    
    See Also:
    
    Constant Field Values
  - VAL_RIGHT
```
public static final java.lang.String VAL_RIGHT
```
    The discrete attribtue value for the tiger being behind the right door
    
    See Also:
    
    Constant Field Values
  - HEAR_LEFT
```
public static final java.lang.String HEAR_LEFT
```
    The observation attribute value for hearing the tiger behind the left door.
    
    See Also:
    
    Constant Field Values
  - HEAR_RIGHT
```
public static final java.lang.String HEAR_RIGHT
```
    The observation attribute value for hearing the tiger behind the right door
    
    See Also:
    
    Constant Field Values
  - DOOR_RESET
```
public static final java.lang.String DOOR_RESET
```
    The observation value for when reaching a new pair of doors (occurs after opening a door)
    
    See Also:
    
    Constant Field Values
  - HEAR_NOTHING
```
public static final java.lang.String HEAR_NOTHING
```
    The observation of hearing nothing (occurs when taking the do nothing action)
    
    See Also:
    
    Constant Field Values
  - includeDoNothing
```
protected boolean includeDoNothing
```
    Whether this domain should include the do nothing action or not
  - listenAccuracy
```
protected double listenAccuracy
```
    The probability of hearing accurately where the tiger is
  - correctDoorReward
```
public double correctDoorReward
```
    the reward for opening the correct door
  - wrongDoorReward
```
public double wrongDoorReward
```
    The reward for opening the wrong door
  - listenReward
```
public double listenReward
```
    The reward for listening
  - nothingReward
```
public double nothingReward
```
    The reward for do nothing.
- Constructor Detail
  - TigerDomain
```
public TigerDomain()
```
    Initializes. There will be no "do nothing" action and the listen accuracy will be set to 0.85
  - TigerDomain
```
public TigerDomain(boolean includeDoNothing)
```
    Initializes. The listen accuracy will be set to 0.85
    
    Parameters:
    
    includeDoNothing - if true, then the do nothing action will be included; if false, then it will not be included
  - TigerDomain
```
public TigerDomain(boolean includeDoNothing,
                   double listenAccuracy)
```
    Initializes
    
    Parameters:
    
    includeDoNothing - if true, then the do nothing action will be included; if false, then it will not be included
    
    listenAccuracy - the listen accuracy
- Method Detail
  - isIncludeDoNothing
```
public boolean isIncludeDoNothing()
```
  - setIncludeDoNothing
```
public void setIncludeDoNothing(boolean includeDoNothing)
```
  - getListenAccuracy
```
public double getListenAccuracy()
```
  - setListenAccuracy
```
public void setListenAccuracy(double listenAccuracy)
```
  - getCorrectDoorReward
```
public double getCorrectDoorReward()
```
  - setCorrectDoorReward
```
public void setCorrectDoorReward(double correctDoorReward)
```
  - getWrongDoorReward
```
public double getWrongDoorReward()
```
  - setWrongDoorReward
```
public void setWrongDoorReward(double wrongDoorReward)
```
  - getListenReward
```
public double getListenReward()
```
  - setListenReward
```
public void setListenReward(double listenReward)
```
  - getNothingReward
```
public double getNothingReward()
```
  - setNothingReward
```
public void setNothingReward(double nothingReward)
```
  - generateDomain
```
public Domain generateDomain()
```
    Description copied from interface: DomainGenerator
    
    Returns a newly instanced Domain object
    
    Specified by:
    
    generateDomain in interface DomainGenerator
    
    Returns:
    
    the newly instantiated Domain object.
  - randomSideStateGenerator
```
public static StateGenerator randomSideStateGenerator()
```
    Returns a StateGenerator that 50% of the time generates an hidden tiger state with the tiger on the left side, and 50% time on the right.
    
    Returns:
    
    a StateGenerator
  - randomSideStateGenerator
```
public static StateGenerator randomSideStateGenerator(double probLeft)
```
    Returns a StateGenerator that some of the of the time generates an hidden tiger state with the tiger on the left side, and others on the right. Probability of left side is specified with the argument probLeft
    
    Parameters:
    
    probLeft - the probability that a state with the tiger on the left side will be generated
    
    Returns:
    
    a StateGenerator
  - getInitialBeliefState
```
public static TabularBeliefState getInitialBeliefState(PODomain domain)
```
    Generates an initial TabularBeliefState in which the it is equally uncertain where the tiger is (50/50).
    
    Parameters:
    
    domain - the domain
    
    Returns:
    
    an initial TabularBeliefState in which the it is equally uncertain where the tiger is (50/50).
  - main
```
public static void main(java.lang.String[] args)
```
    Main method for interacting with the tiger domain via an EnvironmentShell By default, the TerminalExplorer interacts with the partially observable environment (SimulatedPOEnvironment), which means you only get to see the observations that the agent would. However, if you set the first command-line argument to be "h", then the explorer will explorer the underlying fully observable MDP states.
    
    Parameters:
    
    args - either empty or ["h"]; provide "h" to explorer the underlying fully observable tiger MDP.

Class TigerDomain

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

VAR_DOOR

VAR_HEAR

ACTION_LEFT

ACTION_RIGHT

ACTION_LISTEN

ACTION_DO_NOTHING

VAL_LEFT

VAL_RIGHT

HEAR_LEFT

HEAR_RIGHT

DOOR_RESET

HEAR_NOTHING

includeDoNothing

listenAccuracy

correctDoorReward

wrongDoorReward

listenReward

nothingReward

Constructor Detail

TigerDomain

TigerDomain

TigerDomain

Method Detail

isIncludeDoNothing

setIncludeDoNothing

getListenAccuracy

setListenAccuracy

getCorrectDoorReward

setCorrectDoorReward

getWrongDoorReward

setWrongDoorReward

getListenReward

setListenReward

getNothingReward

setNothingReward

generateDomain

randomSideStateGenerator

randomSideStateGenerator

getInitialBeliefState

main