DUOL (Java Statistical Analysis Tool 0.0.8 API)

java.lang.Object
- jsat.classifiers.BaseUpdateableClassifier
- - jsat.classifiers.linear.kernelized.DUOL

All Implemented Interfaces:

Serializable, Cloneable, BinaryScoreClassifier, Classifier, UpdateableClassifier, Parameterized
```
public class DUOL
extends BaseUpdateableClassifier
implements BinaryScoreClassifier, Parameterized
```
Provides an implementation of Double Update Online Learning (DUOL) algorithm. It is a kernelized extension of PassiveAggressive (PA-I) where one previously learned support vector may be updated upon each addition to the support vector set. The SV set is unbounded in size. The objective function is not identical because of the dual updates.

Using a larger C value for DUOL has theoretical improvements, as it increases the number of "strong" dual updates. The default value is set to 10 as suggested in the paper. See:
- Zhao, P., Hoi, S. C. H.,&Jin, R. (2011). Double Updating Online Learning. Journal of Machine Learning Research, 12, 1587–1615. Retrieved from here
- Zhao, P., Hoi, S. C. H.,&Jin, R. (2009). DUOL: A Double Updating Approach for Online Learning. In Y. Bengio, D. Schuurmans, J. Lafferty, C. K. I. Williams,&A. Culotta (Eds.), Advances in Neural Information Processing Systems 22 (pp. 2259–2267).
Author:

Edward Raff

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected List<Double>`	`accelCache`
`protected List<Double>`	`alphas` Signed weights for each support vector.
`protected double`	`C`
`protected List<Double>`	`f_s` Cached outputs of the current decision function on each support vector
`protected KernelTrick`	`k` Kernel trick to use
`protected DoubleList`	`kTmp` Stores the values of k(x_i, y) for reuse when observing a new example
`protected double`	`rho`
`protected List<Vec>`	`S` Set of support vectors

Constructor Summary

Constructors
Modifier Constructor and Description

protected DUOL(DUOL other)
Copy constructor

DUOL(KernelTrick k)
Creates a new DUOL learner

Constructors
Modifier	Constructor and Description
`protected`	`DUOL(DUOL other)` Copy constructor
	`DUOL(KernelTrick k)` Creates a new DUOL learner

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`CategoricalResults`	`classify(DataPoint data)` Performs classification on the given data point.
`DUOL`	`clone()`
`double`	`getC()` Returns the aggressiveness parameter
`KernelTrick`	`getKernel()` Returns the kernel trick in use
`Parameter`	`getParameter(String paramName)` Returns the parameter with the given name.
`List<Parameter>`	`getParameters()` Returns the list of parameters that can be altered for this learner.
`double`	`getRho()` Returns the "conflict" parameter value for the threshold of performing double updates
`double`	`getScore(DataPoint dp)` Returns the numeric score for predicting a class of a given data point, where the sign of the value indicates which class the data point is predicted to belong to.
`static Distribution`	`guessC(DataSet d)` Guesses the distribution to use for the C parameter
`void`	`setC(double C)` Sets the aggressiveness parameter.
`void`	`setKernel(KernelTrick k)` Sets the kernel trick to use
`void`	`setRho(double rho)` Sets the "conflict" parameter, which controls how often double updates are performed.
`void`	`setUp(CategoricalData[] categoricalAttributes, int numericAttributes, CategoricalData predicting)` Prepares the classifier to begin learning from its `UpdateableClassifier.update(jsat.classifiers.DataPoint, int)` method.
`boolean`	`supportsWeightedData()` Indicates whether the model knows how to train using weighted data points.
`void`	`update(DataPoint dataPoint, int targetClass)` Updates the classifier by giving it a new data point to learn from.

Methods inherited from class jsat.classifiers.BaseUpdateableClassifier
getEpochs, setEpochs, trainC, trainC, trainEpochs

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

Methods inherited from interface jsat.classifiers.Classifier
trainC, trainC

- Field Detail
  - k
```
protected KernelTrick k
```
    Kernel trick to use
  - S
```
protected List<Vec> S
```
    Set of support vectors
  - f_s
```
protected List<Double> f_s
```
    Cached outputs of the current decision function on each support vector
  - alphas
```
protected List<Double> alphas
```
    Signed weights for each support vector. Original paper uses the notation of gamma in the original paper, but this is weird and easily confused with the sign values y_i. The sign class values can be obtained from the signed alphas using Math.signum(double)
  - accelCache
```
protected List<Double> accelCache
```
  - kTmp
```
protected DoubleList kTmp
```
    Stores the values of k(x_i, y) for reuse when observing a new example
  - rho
```
protected double rho
```
  - C
```
protected double C
```
- Constructor Detail
  - DUOL
```
public DUOL(KernelTrick k)
```
    Creates a new DUOL learner
    
    Parameters:
    
    k - the kernel to use
  - DUOL
```
protected DUOL(DUOL other)
```
    Copy constructor
    
    Parameters:
    
    other - the object to copy
- Method Detail
  - clone
```
public DUOL clone()
```
    Specified by:
    
    clone in interface BinaryScoreClassifier
    
    Specified by:
    
    clone in interface Classifier
    
    Specified by:
    
    clone in interface UpdateableClassifier
    
    Specified by:
    
    clone in class BaseUpdateableClassifier
  - setC
```
public void setC(double C)
```
    Sets the aggressiveness parameter. Increasing the value of this parameter increases the aggressiveness of the algorithm. It must be a positive value. This parameter essentially performs a type of regularization on the updates
    
    Parameters:
    
    C - the aggressiveness parameter in (0, Inf)
  - getC
```
public double getC()
```
    Returns the aggressiveness parameter
    
    Returns:
    
    the aggressiveness parameter
  - setRho
```
public void setRho(double rho)
```
    Sets the "conflict" parameter, which controls how often double updates are performed. Smaller (near zero) values tend to produce more double updates, with values near 1 producing few double updates. The value must be in the range [0, 1]
    
    Parameters:
    
    rho - the conflict parameter for when to update a second support vector
  - getRho
```
public double getRho()
```
    Returns the "conflict" parameter value for the threshold of performing double updates
    
    Returns:
    
    the "conflict" parameter value for the threshold of performing double updates
  - setKernel
```
public void setKernel(KernelTrick k)
```
    Sets the kernel trick to use
    
    Parameters:
    
    k - the kernel trick to use
  - getKernel
```
public KernelTrick getKernel()
```
    Returns the kernel trick in use
    
    Returns:
    
    the kernel trick in use
  - setUp
```
public void setUp(CategoricalData[] categoricalAttributes,
                  int numericAttributes,
                  CategoricalData predicting)
```
    Description copied from interface: UpdateableClassifier
    
    Prepares the classifier to begin learning from its UpdateableClassifier.update(jsat.classifiers.DataPoint, int) method.
    
    Specified by:
    
    setUp in interface UpdateableClassifier
    
    Parameters:
    
    categoricalAttributes - an array containing the categorical attributes that will be in each data point
    
    numericAttributes - the number of numeric attributes that will be in each data point
    
    predicting - the information for the target class that will be predicted
  - update
```
public void update(DataPoint dataPoint,
                   int targetClass)
```
    Description copied from interface: UpdateableClassifier
    
    Updates the classifier by giving it a new data point to learn from.
    
    Specified by:
    
    update in interface UpdateableClassifier
    
    Parameters:
    
    dataPoint - the data point to learn
    
    targetClass - the target class of the data point
  - classify
```
public CategoricalResults classify(DataPoint data)
```
    Description copied from interface: Classifier
    
    Performs classification on the given data point.
    
    Specified by:
    
    classify in interface Classifier
    
    Parameters:
    
    data - the data point to classify
    
    Returns:
    
    the results of the classification.
  - getScore
```
public double getScore(DataPoint dp)
```
    Description copied from interface: BinaryScoreClassifier
    
    Returns the numeric score for predicting a class of a given data point, where the sign of the value indicates which class the data point is predicted to belong to.
    
    Specified by:
    
    getScore in interface BinaryScoreClassifier
    
    Parameters:
    
    dp - the data point to predict the class label of
    
    Returns:
    
    the score for the given data point
  - supportsWeightedData
```
public boolean supportsWeightedData()
```
    Description copied from interface: Classifier
    
    Indicates whether the model knows how to train using weighted data points. If it does, the model will train assuming the weights. The values returned by this method may change depending on the parameters set for the model.
    
    Specified by:
    
    supportsWeightedData in interface Classifier
    
    Returns:
    
    true if the model supports weighted data, false otherwise
  - guessC
```
public static Distribution guessC(DataSet d)
```
    Guesses the distribution to use for the C parameter
    
    Parameters:
    
    d - the dataset to get the guess for
    
    Returns:
    
    the guess for the C parameter
    
    See Also:
    
    setC(double)
  - getParameters
```
public List<Parameter> getParameters()
```
    Description copied from interface: Parameterized
    
    Returns the list of parameters that can be altered for this learner.
    
    Specified by:
    
    getParameters in interface Parameterized
    
    Returns:
    
    the list of parameters that can be altered for this learner.
  - getParameter
```
public Parameter getParameter(String paramName)
```
    Description copied from interface: Parameterized
    
    Returns the parameter with the given name. Two different strings may map to a single Parameter object. An ASCII only string, and a Unicode style string.
    
    Specified by:
    
    getParameter in interface Parameterized
    
    Parameters:
    
    paramName - the name of the parameter to obtain
    
    Returns:
    
    the Parameter in question, or null if no such named Parameter exists.

Class DUOL

Field Summary

Constructor Summary

Method Summary

Methods inherited from class jsat.classifiers.BaseUpdateableClassifier

Methods inherited from class java.lang.Object

Methods inherited from interface jsat.classifiers.Classifier

Field Detail

k

S

f_s

alphas

accelCache

kTmp

rho

C

Constructor Detail

DUOL

DUOL

Method Detail

clone

setC

getC

setRho

getRho

setKernel

getKernel

setUp

update

classify

getScore

supportsWeightedData

guessC

getParameters

getParameter