Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||Efficient orthogonal non-negative matrix factorization over stiefel manifold|
|Citation:||Proceedings of the 25th ACM International Conference on Information and Knowledge Management (CIKM '16), 2016 / vol.24-28-October-2016, pp.1743-1752|
|Publisher:||Association for Computing Machinery (ACM)|
|Conference Name:||ACM International Conference on Information and Knowledge Management (CIKM '16) (24 Oct 2016 - 28 Oct 2016 : Indianapolis, IN, USA)|
|Wei Emma Zhang, Mingkui Tan, Quan Z. Sheng, Lina Yao, Qingfeng Shi|
|Abstract:||Orthogonal Non-negative Matrix Factorization (ONMF) ap- proximates a data matrix X by the product of two lower- dimensional factor matrices: X ≈ UVT, with one of them orthogonal. ONMF has been widely applied for clustering, but it often suffers from high computational cost due to the orthogonality constraint. In this paper, we propose a method, called Nonlinear Riemannian Conjugate Gradient ONMF (NRCG-ONMF), which updates U and V alterna- tively and preserves the orthogonality of U while achiev- ing fast convergence speed. Specifically, in order to update U, we develop a Nonlinear Riemannian Conjugate Gradi- ent (NRCG) method on the Stiefel manifold using Barzilai- Borwein (BB) step size. For updating V, we use a closed- form solution under non-negativity constraint. Extensive experiments on both synthetic and real-world data sets show consistent superiority of our method over other approaches in terms of orthogonality preservation, convergence speed and clustering performance.|
|Keywords:||Orthogonal NMF; Stiefel Manifold; Clustering|
|Rights:||© 2016 ACM|
|Appears in Collections:||Computer Science publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.