Abstract | ||
---|---|---|
In this paper, we present several descent methods that can be ap- plied to nonnegative matrix factorization and we analyze a recently developed fast block coordinate method. We also give a comparison of these different methods and show that the new block coordinate method has better properties in terms of approximation error and com- plexity. By interpreting this method as a rank-one approximation of the residue matrix, we also extend it to the nonnegative tensor factor- ization and introduce some variants of the method by imposing some additional controllable constraints such as: sparsity, discreteness and smoothness. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1007/978-94-007-0602-6_13 | Clinical Orthopaedics and Related Research |
Keywords | Field | DocType |
nonnegative matrix,factorization,algorithm,nonnegative matrix factorization,information retrieval,numerical analysis,approximation error | Discrete mathematics,Mathematical optimization,Nonnegative tensor factorization,Nonnegative matrix,Matrix (mathematics),Euler's factorization method,Incomplete LU factorization,Non-negative matrix factorization,Smoothness,Mathematics,Approximation error | Journal |
Volume | Citations | PageRank |
abs/0801.3 | 23 | 1.27 |
References | Authors | |
10 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ngoc-Diep Ho | 1 | 36 | 2.89 |
Paul van Dooren | 2 | 649 | 90.48 |
Vincent D. Blondel | 3 | 1880 | 184.86 |