Matrix Factorization Task
A Matrix Factorization Task is a decomposition task that requires a matrix factorization structure (that decomposes an [math]m \times n[/math] matrix [math]X[/math] into a product of matrices (in some canonical form) - typically into [math]m×k[/math] and [math]k×n[/math] matrices, where typically [math]k \ll n[/math] and [math]k \ll m[/math]).
- input: a Matrix Structure.
- output: a Matrix Factorization Structure.
- It can be solved by a Matrix Decomposition System (that implements a matrix decomposition algorithm).
- It can range from being a Nonnegative Matrix Factorization to being a Positive Matrix Factorization to being ...
- It can range from being an Exact Matrix Decomposition Task to being an Approximate Matrix Factorization Task.
- It can range from being a Regularized Matrix Decomposition Task to being an Non-Regularized Matrix Factorization Task.
- It can range from being a Weighted Matrix Decomposition Task to being an Unweighted Matrix Factorization Task.
- It can range from being a Boolean Matrix Decomposition Task to being an Integer Matrix Decomposition Task to being Real Matrix Decomposition Task.
- It can range from being a Low-Rank Matrix Factorization Task to being a Large-Rank Matrix Factorization Algorithm.
- See: Lossy Compression, Adjacency Matrix, Latent Factor Algorithm, Linear Algebra Task.
- (Wikipedia, 2015) ⇒ http://en.wikipedia.org/wiki/matrix_decomposition Retrieved:2015-1-17.
- (Skillicorn, 2007) ⇒ David Skillicorn. (2007). “Understanding Complex Datasets: Data Mining with Matrix Decompositions." Chapman & Hall/CRC.
- (Serre, 2002) ⇒ Denis Serre. (2002). “Matrices. Theory and Applications.” In: Grad. Texts in Math, 216.
- (Luo & Hancock, 2001) ⇒ Bin Luo and Edwin R. Hancock. (2001). “Structural Graph Matching Using the EM Algorithm and Singular Value Decomposition.” In: IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(10).