SVDIncompleteIncrementalLearning Class Reference

This class computes SVD using incomplete incremental batch learning, as described in the following paper: More...

Public Member Functions
	SVDIncompleteIncrementalLearning (double u=0.001, double kw=0, double kh=0)
	Initialize the parameters of SVDIncompleteIncrementalLearning. More...
template < typename MatType >
void	HUpdate (const MatType &V, const arma::mat &W, arma::mat &H)
	The update rule for the encoding matrix H. More...
template < typename MatType >
void	Initialize (const MatType &, const size_t)
	Initialize parameters before factorization. More...
template < typename MatType >
void	WUpdate (const MatType &V, arma::mat &W, const arma::mat &H)
	The update rule for the basis matrix W. More...

Detailed Description

This class computes SVD using incomplete incremental batch learning, as described in the following paper:

@techreport{ma2008guide,
  title={A Guide to Singular Value Decomposition for Collaborative
      Filtering},
  author={Ma, Chih-Chao},
  year={2008},
  institution={Department of Computer Science, National Taiwan University}
}

This class implements 'Algorithm 2' as given in the paper. Incremental learning modifies only some feature values in W and H after scanning part of the input matrix (V). This differs from batch learning, which considers every element in V for each update of W and H. The regularization technique is also different: in incomplete incremental learning, regularization takes into account the number of elements in a given column of V.

See also

SVDBatchLearning

Definition at line 43 of file svd_incomplete_incremental_learning.hpp.

Constructor & Destructor Documentation

SVDIncompleteIncrementalLearning()

SVDIncompleteIncrementalLearning ( double  u = 0.001, double  kw = 0, double  kh = 0  )

inline

Initialize the parameters of SVDIncompleteIncrementalLearning.

Parameters

u	Step value used in batch learning.
kw	Regularization constant for W matrix.
kh	Regularization constant for H matrix.

Definition at line 53 of file svd_incomplete_incremental_learning.hpp.

Member Function Documentation

HUpdate()

void HUpdate ( const MatType &  V, const arma::mat &  W, arma::mat &  H  )

inline

The update rule for the encoding matrix H.

The function takes in all the matrices and only changes the value of the H matrix.

Parameters

V	Input matrix to be factorized.
W	Basis matrix.
H	Encoding matrix to be updated.

Definition at line 121 of file svd_incomplete_incremental_learning.hpp.

Initialize()

void Initialize ( const MatType &  , const size_t    )

inline

Initialize parameters before factorization.

This function must be called before a new factorization. This simply sets the column being considered to 0, so the input matrix and rank are not used.

Parameters

*	(dataset) Input matrix to be factorized.
*	(rank) of factorization

Definition at line 70 of file svd_incomplete_incremental_learning.hpp.

WUpdate()

void WUpdate ( const MatType &  V, arma::mat &  W, const arma::mat &  H  )

inline

The update rule for the basis matrix W.

The function takes in all the matrices and only changes the value of the W matrix.

Parameters

V	Input matrix to be factorized.
W	Basis matrix to be updated.
H	Encoding matrix.

Definition at line 86 of file svd_incomplete_incremental_learning.hpp.

---

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

• /home/ryan/src/mlpack.org/_src/mlpack-git/src/mlpack/methods/amf/update_rules/svd_incomplete_incremental_learning.hpp