Package com.google.privacy.differentialprivacy

Class GaussianNoise

java.lang.Object

com.google.privacy.differentialprivacy.GaussianNoise

All Implemented Interfaces:

Noise

public class GaussianNoise
extends java.lang.Object
implements Noise

Generates and adds Gaussian noise to a raw piece of numerical data such that the result is securely differentially private.

The Gaussian noise is generated according to the binomial sampling mechanism described here. This approach is robust against unintentional privacy leaks due to artifacts of floating point arithmetic.

Constructor Summary

Constructors

Constructor	Description
GaussianNoise()	Returns a Noise instance initialized with a secure randomness source.

Method Summary

Modifier and Type	Method	Description
double	addNoise(double x, int l0Sensitivity, double lInfSensitivity, double epsilon, java.lang.Double delta)	Adds Gaussian noise to x such that the output is (epsilon, delta)-differentially private, with respect to the specified L_0 and L_inf sensitivities.
long	addNoise(long x, int l0Sensitivity, long lInfSensitivity, double epsilon, java.lang.Double delta)	Adds Gaussian noise to the integer x such that the output is (epsilon, delta)-differentially private, with respect to the specified L_0 and L_inf sensitivities.
ConfidenceInterval	computeConfidenceInterval(double noisedX, int l0Sensitivity, double lInfSensitivity, double epsilon, java.lang.Double delta, double alpha)	Computes a confidence interval that contains the raw value x passed to addNoise(double, int, double, double, Double) with a probability equal to 1 - alpha based on the specified noisedX and noise parameters.
ConfidenceInterval	computeConfidenceInterval(long noisedX, int l0Sensitivity, long lInfSensitivity, double epsilon, java.lang.Double delta, double alpha)	Computes a confidence interval that contains the raw integer value x passed to addNoise(long, int, long, double, Double) with a probability greater or equal to 1 - alpha based on the specified noisedX and noise parameters.
double	computeQuantile(double rank, double x, int l0Sensitivity, double lInfSensitivity, double epsilon, java.lang.Double delta)	Computes the quantile z satisfying Pr[Y <= z] = rank for a Gaussian random variable Y with mean x and variance according to the specified privacy parameters.
com.google.differentialprivacy.SummaryOuterClass.MechanismType	getMechanismType()

Methods inherited from class java.lang.Object

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

Constructor Detail

GaussianNoise

public GaussianNoise()

Returns a Noise instance initialized with a secure randomness source.

Method Detail

addNoise

public double addNoise(double x,
                       int l0Sensitivity,
                       double lInfSensitivity,
                       double epsilon,
                       java.lang.Double delta)

Adds Gaussian noise to x such that the output is (epsilon, delta)-differentially private, with respect to the specified L_0 and L_inf sensitivities.

Specified by:

addNoise in interface Noise

addNoise

public long addNoise(long x,
                     int l0Sensitivity,
                     long lInfSensitivity,
                     double epsilon,
                     @Nullable
                     java.lang.Double delta)

Adds Gaussian noise to the integer x such that the output is (epsilon, delta)-differentially private, with respect to the specified L_0 and L_inf sensitivities.

Specified by:

addNoise in interface Noise

getMechanismType

public com.google.differentialprivacy.SummaryOuterClass.MechanismType getMechanismType()

Specified by:

getMechanismType in interface Noise

computeConfidenceInterval

public ConfidenceInterval computeConfidenceInterval(double noisedX,
                                                    int l0Sensitivity,
                                                    double lInfSensitivity,
                                                    double epsilon,
                                                    java.lang.Double delta,
                                                    double alpha)

Computes a confidence interval that contains the raw value x passed to addNoise(double, int, double, double, Double) with a probability equal to 1 - alpha based on the specified noisedX and noise parameters.

Refer to this doc for more information.

Specified by:

computeConfidenceInterval in interface Noise

computeConfidenceInterval

public ConfidenceInterval computeConfidenceInterval(long noisedX,
                                                    int l0Sensitivity,
                                                    long lInfSensitivity,
                                                    double epsilon,
                                                    java.lang.Double delta,
                                                    double alpha)

Computes a confidence interval that contains the raw integer value x passed to addNoise(long, int, long, double, Double) with a probability greater or equal to 1 - alpha based on the specified noisedX and noise parameters.

Refer to this doc for more information.

Specified by:

computeConfidenceInterval in interface Noise

computeQuantile

public double computeQuantile(double rank,
                              double x,
                              int l0Sensitivity,
                              double lInfSensitivity,
                              double epsilon,
                              @Nullable
                              java.lang.Double delta)

Computes the quantile z satisfying Pr[Y <= z] = rank for a Gaussian random variable Y with mean x and variance according to the specified privacy parameters.

Specified by:

computeQuantile in interface Noise