Robust informed split gradient NMF using αβ-divergence for source apportionment

Source apportionment is a very challenging topic for which non-negative source separation is well-suited. Recently, we proposed several informed Non-negative Matrix Factorization (NMF) for which some expert knowledge was introduced. These methods were all dealing with some set values of one factor together with the row sum-to-one property by either processing each constraint alternatingly or using a new parameterization which involves all of them. However, this last method was sensitive to the presence of outliers. In this paper, we thus propose a new robust informed Split Gradient NMF method which is based on a weighted αβ-divergence cost function. Experiments conducted for several input SNR with and without outliers on simulated mixtures of particulate matter sources show the relevance of the new approach.

[1]  Matthieu Puigt,et al.  Bound constrained weighted NMF for industrial source apportionment , 2014, 2014 IEEE International Workshop on Machine Learning for Signal Processing (MLSP).

[2]  M. Puigt,et al.  Non-negative Matrix Factorization under equality constraints—a study of industrial source identification , 2014 .

[3]  Nicolas Dobigeon,et al.  Nonlinear Hyperspectral Unmixing With Robust Nonnegative Matrix Factorization , 2014, IEEE Transactions on Image Processing.

[4]  Matthieu Puigt,et al.  Blind mobile sensor calibration using an informed nonnegative matrix factorization with a relaxed rendezvous model , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[5]  Antonio J. Plaza,et al.  A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote Sensing , 2014, IEEE Signal Processing Magazine.

[6]  Bernt Schiele,et al.  Introducing a weighted non-negative matrix factorization for image classification , 2003, Pattern Recognit. Lett..

[7]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[8]  Jérôme Idier,et al.  Algorithms for Nonnegative Matrix Factorization with the β-Divergence , 2010, Neural Computation.

[9]  Jean-Luc Starck,et al.  Sparse and Non-Negative BSS for Noisy Data , 2013, IEEE Transactions on Signal Processing.

[10]  Dan Hu,et al.  Non‐negative Matrix Factorization Algorithms and Applications , 2014 .

[11]  Xuan Li,et al.  Robust Nonnegative Matrix Factorization via Half-Quadratic Minimization , 2012, 2012 IEEE 12th International Conference on Data Mining.

[12]  Cédric Richard,et al.  Split gradient method for nonnegative matrix factorization , 2010, 2010 18th European Signal Processing Conference.

[13]  Emmanuel Vincent,et al.  The 2008 Signal Separation Evaluation Campaign: A Community-Based Approach to Large-Scale Evaluation , 2009, ICA.

[14]  M. Viana,et al.  Inter-comparison of receptor models for PM source apportionment: Case study in an industrial area , 2008 .

[15]  Peter Filzmoser,et al.  Robust Multivariate Methods in Chemometrics , 2020, Comprehensive Chemometrics.

[16]  A. Ben Hamza,et al.  Reconstruction of reflectance spectra using robust nonnegative matrix factorization , 2006, IEEE Transactions on Signal Processing.

[17]  Matthieu Puigt,et al.  Blind Calibration of Mobile Sensors Using Informed Nonnegative Matrix Factorization , 2015, LVA/ICA.

[18]  Cédric Févotte,et al.  Alternating direction method of multipliers for non-negative matrix factorization with the beta-divergence , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[19]  Paul D. Gader,et al.  A Signal Processing Perspective on Hyperspectral Unmixing , 2014 .

[20]  Jérôme Idier,et al.  Algorithms for nonnegative matrix factorization with the beta-divergence , 2010, ArXiv.

[21]  Émilie Chouzenoux,et al.  Algorithme primal-dual de points intérieurs pour l'estimation pénalisée des cartes d'abondances en imagerie hyperspectrale , 2011, Traitement du Signal.

[22]  Chein-I Chang,et al.  Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery , 2001, IEEE Trans. Geosci. Remote. Sens..

[23]  Chris H. Q. Ding,et al.  Robust nonnegative matrix factorization using L21-norm , 2011, CIKM '11.

[24]  Sergio Cruces,et al.  Generalized Alpha-Beta Divergences and Their Application to Robust Nonnegative Matrix Factorization , 2011, Entropy.

[25]  Andrzej Cichocki,et al.  Non-negative matrix factorization with alpha-divergence , 2008, Pattern Recognit. Lett..

[26]  Antoine Abche,et al.  Split Gradient Method for Informed Non-negative Matrix Factorization , 2015, LVA/ICA.