Unsupervised Learning of Overlapping Image Components Using Divisive Input Modulation

This paper demonstrates that nonnegative matrix factorisation is mathematically related to a class of neural networks that employ negative feedback as a mechanism of competition. This observation inspires a novel learning algorithm which we call Divisive Input Modulation (DIM). The proposed algorith...

Descripción completa

Detalles Bibliográficos
Autores principales: Spratling, M. W., De Meyer, K., Kompass, R.
Formato: Texto
Lenguaje:English
Publicado: Hindawi Publishing Corporation 2009
Materias:
Research Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2677748/
https://www.ncbi.nlm.nih.gov/pubmed/19424442
http://dx.doi.org/10.1155/2009/381457

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2677748/
https://www.ncbi.nlm.nih.gov/pubmed/19424442
http://dx.doi.org/10.1155/2009/381457

Ejemplares similares