Stability analysis of multiplicative update algorithms for non-negative matrix factorization

Citations Over Time

Abstract

Multiplicative update algorithms have encountered a great success to solve optimization problems with non-negativity constraints, such as the famous non-negative matrix factorization (NMF) and its many variants. However, despite several years of research on the topic, the understanding of their convergence properties is still to be improved. In this paper, we show that Lyapunov's stability theory provides a very enlightening viewpoint on the problem. We prove the stability of supervised NMF and study the more difficult case of unsupervised NMF. Numerical simulations illustrate those theoretical results, and the convergence speed of NMF multiplicative updates is analyzed.

Related Papers

• → CUR+NMF for learning spectral features from large data matrix(2008)10 cited
• → Non-Negative Matrix Factorization for Note Onset Detection of Audio Signals(2006)12 cited
• → Stability analysis of multiplicative update algorithms for non-negative matrix factorization(2011)6 cited
• → Sparsity promoted non-negative matrix factorization for source separation and detection(2014)3 cited
• → PHASL-NMF: Hierarchical ALS Based Power Non-Negative Matrix Factorization(2023)