Stochastic Gradient Descent (SGD) is a popular online algorithm for large-scale matrix factorization. However, SGD can often be di cult to use for practitioners, because its performance is very sensitive to the choice of the learning rate parameter. In this paper, we present non-negative passiveaggressive (NN-PA), a family of online algorithms for non-negative matrix factorization (NMF). Our al...

Non-negative matrix factorization (NMF) has become a popular technique for finding low-dimensional representations of data. While the standard NMF can only be performed in the original feature space, one variant of NMF, named concept factorization, can be naturally kernelized and inherits all the strengths of NMF. To make use of label information, we propose a semi-supervised concept this paper...

In applications such as speech and audio denoising, music transcription, music and audio based forensics, it is desirable to decompose a single-channel recording into its respective sources, commonly referred to as blind source separation (BSS). One of the techniques used in BSS is non-negative matrix factorization (NMF). In NMF both supervised and unsupervised mode of operations is used. Among...

Abstract In image compression and feature extraction, linear expansions are standardly used. It was pointed out by Lee and Seung that the positivity or non-negativity of a linear expansion is a very powerful constraint, that seems to lead to sparse representations for the images. Their technique, called Non-negative Matrix Factorization (NMF), was shown to be useful in approximating high dimens...

In this paper, we apply a non-negative matrix factorization (NMF) technique to propose a method of estimating noise occurring in non-stationary environments. In the proposed method, the basis matrix of the target noise is first obtained via NMF training. The noise basis is then applied to estimate an activation matrix of the target noise from the noisy signal. The proposed method is finally app...

We propose a convolutive non-negative matrix factorization method to improve the intelligibility of speech signal in the context of adverse noise environment. The noise bases are prior learned with Non-negative Matrix Factorization (NMF) algorithm. A modified convolutive NMF with sparse constraint is then derived to extract speech bases from noisy speech. The divergence function is selected as ...

In this paper, we described our approaches to the RealTime Filtering Task in the TREC 2015 Microblog track. We submitted the results for scenario B: periodic email digest. In this ad hoc search task, we introduced a real-time filtering framework using non-negative matrix factorization. To build this framework, we firstly considered the Real-Time Filtering Task as a short text retrieval problem,...

In this paper, a novel method for facial representation called Spatially Confined Non-Negative Matrix Factorization (SFNMF) is presented. SFNMF aims to extract more spatially confined, parts-based representation from the NMF based representation by merely removing non-prominent region, and focalize on the salient feature. SFNMF derived a significant set of basis which allows a non-subtractive r...

Stochastic matrix factorization (SMF) has proposed and it can be understood as a restriction to non-negative matrix factorization (NMF). SMF is useful for inference of topic models, NMF for binary matrices data, and Bayesian Network. However, it needs some strong assumption to reach unique factorization in SMF and also theoretical prediction accuracy has not yet clarified. In this paper, we stu...

In this paper, we described our approaches to the RealTime Filtering Task in the TREC 2015 Microblog track. We submitted the results for scenario B: periodic email digest. In this ad hoc search task, we introduced a real-time filtering framework using non-negative matrix factorization. To build this framework, we firstly considered the Real-Time Filtering Task as a short text retrieval problem,...