When Does Non-Negative Matrix Factorization Give a Correct Decomposition into Parts?

Part of Advances in Neural Information Processing Systems 16 (NIPS 2003)

Bibtex Metadata Paper


David Donoho, Victoria Stodden


We interpret non-negative matrix factorization geometrically, as the problem of finding a simplicial cone which contains a cloud of data points and which is contained in the positive orthant. We show that under certain conditions, basically requiring that some of the data are spread across the faces of the positive orthant, there is a unique such simpli- cial cone. We give examples of synthetic image articulation databases which obey these conditions; these require separated support and facto- rial sampling. For such databases there is a generative model in terms of ‘parts’ and NMF correctly identifies the ‘parts’. We show that our theoretical results are predictive of the performance of published NMF code, by running the published algorithms on one of our synthetic image articulation databases.