Learning vector quantization and relevances in complex coefficient space

In this contribution, we consider the classification of time series and similar functional data which can be represented in complex Fourier and wavelet coefficient space. We apply versions of learning vector quantization (LVQ) which are suitable for complex-valued data, based on the so-called Wirtinger calculus. It allows for the formulation of gradient-based update rules in the framework of cost-function-based generalized matrix relevance LVQ (GMLVQ). Alternatively, we consider the concatenation of real and imaginary parts of Fourier coefficients in a real-valued feature vector and the classification of time-domain representations by means of conventional GMLVQ. In addition, we consider the application of the method in combination with wavelet-space features to heartbeat classification.

[1]  R. Haddad,et al.  Multiresolution Signal Decomposition: Transforms, Subbands, and Wavelets , 1992 .

[2]  Michael Biehl,et al.  Relevance matrices in LVQ , 2007, ESANN.

[3]  Thomas Villmann,et al.  Complex Variants of GLVQ Based on Wirtinger's Calculus , 2016, WSOM.

[4]  John H. Karl,et al.  The discrete Fourier transform , 2010, Advances in Imaging and Electron Physics.

[5]  E. Merenyi,et al.  Relevance-based Feature Extraction from Hyperspectral Images in the Complex Wavelet Domain , 2006, 2006 IEEE Mountain Workshop on Adaptive and Learning Systems.

[6]  Michael Biehl,et al.  Functional approximation for the classification of smooth time series , 2016 .

[7]  Michael Biehl,et al.  Functional Representation of Prototypes in LVQ and Relevance Learning , 2016, WSOM.

[8]  Michael Biehl,et al.  Advanced metric adaptation in Generalized LVQ for classification of mass spectrometry data , 2007 .

[9]  Tim Oates,et al.  Time series classification from scratch with deep neural networks: A strong baseline , 2016, 2017 International Joint Conference on Neural Networks (IJCNN).

[10]  Jianling Sun,et al.  Piecewise Factorization for Time Series Classification , 2015, IC3K.

[11]  Moncef Gabbouj,et al.  A Generic and Robust System for Automated Patient-Specific Classification of ECG Signals , 2009, IEEE Transactions on Biomedical Engineering.

[12]  Thomas Villmann,et al.  Prototype-based models in machine learning. , 2016, Wiley interdisciplinary reviews. Cognitive science.

[13]  Thomas Villmann,et al.  Distance Measures for Prototype Based Classification , 2013, BrainComp.

[14]  Hans-Georg Müller,et al.  Functional Data Analysis , 2016 .

[15]  Atsushi Sato,et al.  Generalized Learning Vector Quantization , 1995, NIPS.

[16]  Michael Biehl,et al.  Adaptive learning for complex-valued data , 2012, ESANN.

[17]  Jan L. G. Dietz,et al.  Knowledge Discovery, Knowledge Engineering and Knowledge Management , 2012, Communications in Computer and Information Science.

[18]  Michael Biehl,et al.  Adaptive Relevance Matrices in Learning Vector Quantization , 2009, Neural Computation.

[19]  Thomas Villmann,et al.  Prototypes and matrix relevance learning in complex fourier space , 2017, 2017 12th International Workshop on Self-Organizing Maps and Learning Vector Quantization, Clustering and Data Visualization (WSOM).

[20]  Manu Thomas,et al.  Classification of cardiac arrhythmias based on dual tree complex wavelet transform , 2014, 2014 International Conference on Communication and Signal Processing.

[21]  Nick Kingsbury,et al.  The dual-tree complex wavelet transform: a new technique for shift invariance and directional filters , 1998 .

[22]  Thomas Villmann,et al.  Generalized functional relevance learning vector quantization , 2011, ESANN.

[23]  Samit Ari,et al.  Patient-specific ECG beat classification technique. , 2014, Healthcare technology letters.

[24]  W. Wirtinger.  Zur formalen Theorie der Funktionen von mehr komplexen Veränderlichen , 1927 .

[25]  Alfred Mertins,et al.  Signal Analysis: Wavelets, Filter Banks, Time-Frequency Transforms and Applications , 1999 .

[26]  Michael Biehl,et al.  Waypoint averaging and step size control in learning by gradient descent (technical report) , 2011 .