edu.princeton.repeatedgames.rgsolve
Class RGSolution

java.lang.Object
  edu.princeton.repeatedgames.rgsolve.RGSolution

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

RGSolution.RGSolutionUserRule

public class RGSolution
extends java.lang.Object
implements java.io.Serializable

This data structure represents rgsolve's solution to a game (including detailed information on the algorithm's steps).

See Also:

Serialized Form

Nested Class Summary

static class	RGSolution.RGSolutionUserRule A special case of RGSolution, for games where the user entered Java-code payoff rules into the rgsolve GUI
static class	RGSolution.TEXT_OUTPUT_TYPE enum for formatting of V* as text table

Field Summary

AlgoParameters.ALGO_TYPE	algo_type algorithm used, AS or APS
int[]	BR1 Best response actions for player 1
int[]	BR2 Best response actions for player 2
boolean	converged did the algorithm converge to the tolerance set in params?
double	error max sup-distance between corresponding vertices of last two iterates
static java.lang.String	extension File extension for serialization: .rgsoln
boolean	fromInnerApproximation whether or not this solution was generated via an inner approximation
Game	game the game associated with this solution
long	genPtsCount the number of potential extreme points generated over course of algoritm
java.util.ArrayList<RGIter>	iterList A list of RGIter objects, each of which stores information on the algorithm's steps within an iteration
int	iters the number of iterations
boolean	keptActions did we store information on action-wise IC-intersections?
AlgoParameters	params The parameters used in constructing this sol'n
Point	punishment_star The equilibrium threat point, (or threat point where algo stopped)
private static long	serialVersionUID Serialization ID
long	time_ms the time (in ms) to convergence
GameExtremePoint[]	V_star The equilibrium payoff set V* (or final payoff set before algorithm terminated.
Point[]	W0 the initial feasible set containing V* used at start of algorithm

Constructor Summary

RGSolution(AlgoParameters.ALGO_TYPE algo_type, Point[] W0, GameExtremePoint[] V_star, Point punishment_star, Game game, int[] BR1, int[] BR2, int iters, long time_ms, boolean converged, double error, java.util.ArrayList<RGIter> iterList, int genPtsCount, boolean keptActions, boolean fromInnerApproximation, AlgoParameters params)
Default constructor for initializing fields of data structure

Method Summary

java.lang.String

getSolutionAsText(RGSolution.TEXT_OUTPUT_TYPE type, int digits)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

serialVersionUID

private static final long serialVersionUID

Serialization ID

See Also:

Constant Field Values

extension

public static final java.lang.String extension

File extension for serialization: .rgsoln

See Also:

Constant Field Values

params

public AlgoParameters params

The parameters used in constructing this sol'n

algo_type

public AlgoParameters.ALGO_TYPE algo_type

algorithm used, AS or APS

game

public Game game

the game associated with this solution

W0

public Point[] W0

the initial feasible set containing V* used at start of algorithm

V_star

public GameExtremePoint[] V_star

The equilibrium payoff set V* (or final payoff set before algorithm terminated.

punishment_star

public Point punishment_star

The equilibrium threat point, (or threat point where algo stopped)

iterList

public java.util.ArrayList<RGIter> iterList

A list of RGIter objects, each of which stores information on the algorithm's steps within an iteration

BR1

public int[] BR1

Best response actions for player 1

BR2

public int[] BR2

Best response actions for player 2

iters

public int iters

the number of iterations

time_ms

public long time_ms

the time (in ms) to convergence

converged

public boolean converged

did the algorithm converge to the tolerance set in params?

error

public double error

max sup-distance between corresponding vertices of last two iterates

genPtsCount

public long genPtsCount

the number of potential extreme points generated over course of algoritm

keptActions

public boolean keptActions

did we store information on action-wise IC-intersections?

fromInnerApproximation

public boolean fromInnerApproximation

whether or not this solution was generated via an inner approximation

Constructor Detail

RGSolution

public RGSolution(AlgoParameters.ALGO_TYPE algo_type,
                  Point[] W0,
                  GameExtremePoint[] V_star,
                  Point punishment_star,
                  Game game,
                  int[] BR1,
                  int[] BR2,
                  int iters,
                  long time_ms,
                  boolean converged,
                  double error,
                  java.util.ArrayList<RGIter> iterList,
                  int genPtsCount,
                  boolean keptActions,
                  boolean fromInnerApproximation,
                  AlgoParameters params)

Default constructor for initializing fields of data structure

Parameters:

algo_type -

W0 -

V_star -

punishment_star -

game -

BR1 -

BR2 -

iters -

time_ms -

converged -

error -

iterList -

genPtsCount -

keptActions -

fromInnerApproximation -

params -

Method Detail

getSolutionAsText

public java.lang.String getSolutionAsText(RGSolution.TEXT_OUTPUT_TYPE type,
                                          int digits)

Parameters:

type - how the text-table should be formatted

digits - the number of output digits

Returns:

A textual representation of the solution V* as a table