Fabian Theis

Andrzej Cichocki

Arie Yeredor

Michael Zibulevsky

This book constitutes the proceedings of the 10th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2012, held in Tel Aviv, Israel, in March 2012. The 20 revised full papers presented together with 42 revised poster papers, 1 keynote lecture, and 2 overview papers for the regular, as well as for the special session were carefully reviewed and selected from numerous submissions. Topics addressed are ranging from theoretical issues such as causality analysis and measures, through novel methods for employing the well-established concepts of sparsity and non-negativity for matrix and tensor factorization, down to a variety of related applications ranging from audio and biomedical signals to precipitation analysis.

10th International Conference, LVA/ICA 2012, Tel Aviv, Israel, March 12-15, 2012, Proceedings

Latent Variable Analysis and Signal Separation

Shun-ichi Amari

Rafal Zdunek

Anh Huy Phan

This book provides a broad survey of models and efficient algorithms for Nonnegative Matrix Factorization (NMF). This includes NMF’s various extensions and modifications, especially Nonnegative Tensor Factorizations (NTF) and Nonnegative Tucker Decompositions (NTD). NMF/NTF and their extensions are increasingly used as tools in signal and image processing, and data analysis, having garnered interest due to their capability to provide new insights and relevant information about the complex latent relationships in experimental data sets. It is suggested that NMF can provide meaningful components with physical interpretations; for example, in bioinformatics, NMF and its extensions have been successfully applied to gene expression, sequence analysis, the functional characterization of genes, clustering and text mining. As such, the authors focus on the algorithms that are most useful in practice, looking at the fastest, most robust, and suitable for large-scale models. Key features: Acts as a single source reference guide to NMF, collating information that is widely dispersed in current literature, including the authors’ own recently developed techniques in the subject area. Uses generalized cost functions such as Bregman, Alpha and Beta divergences, to present practical implementations of several types of robust algorithms, in particular Multiplicative, Alternating Least Squares, Projected Gradient and Quasi Newton algorithms. Provides a comparative analysis of the different methods in order to identify approximation error and complexity. Includes pseudo codes and optimized MATLAB source codes for almost all algorithms presented in the book. The increasing interest in nonnegative matrix and tensor factorizations, as well as decompositions and sparse representation of data, will ensure that this book is essential reading for engineers, scientists, researchers, industry practitioners and graduate students across signal and image processing; neuroscience; data mining and data analysis; computer science; bioinformatics; speech processing; biomedical engineering; and multimedia.

Applications to Exploratory Multi-way Data Analysis and Blind Source Separation

Nonnegative Matrix and Tensor Factorizations

Rolf Unbehauen

The purpose of this book is to present analysis and design principles, procedures and techniques of analog integrated circuits which are to be implemented in MOS (metal oxide semiconductor) technology. MOS technology is becoming dominant in the realization of digital systems, and its use for analog circuits opens new pos sibilities for the design of complex mixed analog/digital VLSI (very large scale in tegration) chips. Although we are focusing attention in this book principally on circuits and systems which can be implemented in CMOS technology, many con siderations and structures are of a general nature and can be adapted to other promising and emerging technologies, namely GaAs (Gallium Arsenide) and BI MOS (bipolar MOS, i. e. circuits which combine both bipolar and CMOS devices) technology. Moreover, some of the structures and circuits described in this book can also be useful without integration. In this book we describe two large classes of analog integrated circuits: • switched capacitor (SC) networks, • continuous-time CMOS (unswitched) circuits. SC networks are sampled-data systems in which electric charges are transferred from one point to another at regular discrete intervals of time and thus the signal samples are stored and processed. Other circuits belonging to this class of sampled-data systems are charge transfer devices (CTD) and charge coupled dev ices (CCD). In contrast to SC circuits, continuous-time CMOS circuits operate continuously in time. They can be considered as subcircuits or building blocks (e. g.

Analysis and Design

MOS Switched-Capacitor and Continuous-Time Integrated Circuits and Systems

With solid theoretical foundations and numerous potential applications, Blind Signal Processing (BSP) is one of the hottest emerging areas in Signal Processing. This volume unifies and extends the theories of adaptive blind signal and image processing and provides practical and efficient algorithms for blind source separation: Independent, Principal, Minor Component Analysis, and Multichannel Blind Deconvolution (MBD) and Equalization. Containing over 1400 references and mathematical expressions Adaptive Blind Signal and Image Processing delivers an unprecedented collection of useful techniques for adaptive blind signal/image separation, extraction, decomposition and filtering of multi-variable signals and data. Offers a broad coverage of blind signal processing techniques and algorithms both from a theoretical and practical point of view Presents more than 50 simple algorithms that can be easily modified to suit the reader's specific real world problems Provides a guide to fundamental mathematics of multi-input, multi-output and multi-sensory systems Includes illustrative worked examples, computer simulations, tables, detailed graphs and conceptual models within self contained chapters to assist self study Accompanying CD-ROM features an electronic, interactive version of the book with fully coloured figures and text. C and MATLAB user-friendly software packages are also provided MATLAB is a registered trademark of The MathWorks, Inc. By providing a detailed introduction to BSP, as well as presenting new results and recent developments, this informative and inspiring work will appeal to researchers, postgraduate students, engineers and scientists working in biomedical engineering, communications, electronics, computer science, optimisations, finance, geophysics and neural networks.

Learning Algorithms and Applications

Adaptive Blind Signal and Image Processing

Andrzej Cichocki

About

Books

Join the club