Sequential Sparse NMF
نویسندگان
چکیده
Nonnegative Matrix Factorization (NMF) is a standard tool for data analysis. An important variant is the Sparse NMF problem. A natural measure of sparsity is the L0 norm, however its optimization is NP-hard. Here, we consider a sparsity measure linear in the ratio of the L1 and L2 norms, and propose an efficient algorithm to handle the norm constraints which arise when optimizing this measure. Although algorithms for solving these are available, they are typically inefficient. We present experimental evidence that our new algorithm performs an order of magnitude faster compared to the previous state-ofthe-art.
منابع مشابه
Learning Sparse Representations by Non-Negative Matrix Factorization and Sequential Cone Programming
We exploit the biconvex nature of the Euclidean non-negative matrix factorization (NMF) optimization problem to derive optimization schemes based on sequential quadratic and second order cone programming. We show that for ordinary NMF, our approach performs as well as existing stateof-the-art algorithms, while for sparsity-constrained NMF, as recently proposed by P. O. Hoyer in JMLR 5 (2004), i...
متن کاملIterative Weighted Non-smooth Non-negative Matrix Factorization for Face Recognition
Non-negative Matrix Factorization (NMF) is a part-based image representation method. It comes from the intuitive idea that entire face image can be constructed by combining several parts. In this paper, we propose a framework for face recognition by finding localized, part-based representations, denoted “Iterative weighted non-smooth non-negative matrix factorization” (IWNS-NMF). A new cost fun...
متن کاملFast Local Algorithms for Large Scale Nonnegative Matrix and Tensor Factorizations
Nonnegative matrix factorization (NMF) and its extensions such as Nonnegative Tensor Factorization (NTF) have become prominent techniques for blind sources separation (BSS), analysis of image databases, data mining and other information retrieval and clustering applications. In this paper we propose a family of efficient algorithms for NMF/NTF, as well as sparse nonnegative coding and represent...
متن کاملAn iterative hard thresholding approach to ℓ0 sparse Hellinger NMF
Performance of Non-negative Matrix Factorisation (NMF) can be diminished when the underlying factors consist of elements that overlap in the matrix to be factorised. The use of `0 sparsity may improve NMF, however such approaches are generally limited to Euclidean distance. We have previously proposed a stepwise `0 method for Hellinger distance, leading to improved sparse NMF. We extend sparse ...
متن کاملSparse Nonnegative Matrix Factorization for Clustering
Properties of Nonnegative Matrix Factorization (NMF) as a clustering method are studied by relating its formulation to other methods such as K-means clustering. We show how interpreting the objective function of K-means as that of a lower rank approximation with special constraints allows comparisons between the constraints of NMF and K-means and provides the insight that some constraints can b...
متن کامل