Ghostbusters: A Parts-based NMF Algorithm
2013 (English)In: IEEE Irish Signals and Systems Conference / [ed] IET, IEEE, IET, 2013, 1-8 p.Conference paper (Refereed)
An exact nonnegative matrix decomposition algorithm is proposed. This is achieved by 1) Taking a nonlinear approximation of a sparse real-valued dataset at a given tolerance-to-error constraint, c; Choosing an arbitrary lectic ordering on the rows or column entries; And, then systematically applying a closure operator, so that all closures are selected. Assuming a nonnegative hierarchical closure structure (a Galois lattice) ensures the data has a unique ordered overcomplete dictionary representation. Parts-based constraints on these closures can then be used to specify and supervise the form of the solution. We illustrate that this approach outperforms NMF on two standard NMF datasets: it exhibits the properties described above; It is correct and exact.
Place, publisher, year, edition, pages
IET, 2013. 1-8 p.
Nonnegative Matrix Factorization, Lectic Orderings, Unique Solutions
Computer Systems Signal Processing
Research subject Applied and Computational Mathematics
IdentifiersURN: urn:nbn:se:kth:diva-173815DOI: 10.1049/ic.2013.0050OAI: oai:DiVA.org:kth-173815DiVA: diva2:855009
IEEE Irish Signals and Systems Conference
QC 201509212015-09-182015-09-182015-09-21Bibliographically approved