How can we select the best performing data-driven model? How can we rigorously estimate its generalization error? Statistical learning theory answers these questions by deriving non-asymptotic bounds on the generalization error of a model or, in other words, by upper bounding the true error of the learned model based just on quantities computed on the available data. However, for a long time, Statistical learning theory has been considered only an abstract theoretical framework, useful for inspiring new learning approaches, but with limited applicability to practical problems. The purpose of this book is to give an intelligible overview of the problems of model selection and error estimation, by focusing on the ideas behind the different statistical learning theory approaches and simplifying most of the technical aspects with the purpose of making them more accessible and usable in practice. The book starts by presenting the seminal works of the 80’s and includes the most recent results. It discusses open problems and outlines future directions for research.

The classification of patterns is an important area of research which is central to all pattern recognition fields, including speech, image, robotics, and data analysis. Neural networks have been used successfully in a number of these fields, but so far their application has been based on a 'black box approach' with no real understanding of how they work. In this book, Sarunas Raudys - an internationally respected researcher in the area - provides an excellent mathematical and applied introduction to how neural network classifiers work and how they should be used.. .

The 77th Annual International Meeting of the Psychometric Society (IMPS) brought together quantitative researchers who focus on methods relevant to psychology. The conference included workshops, invited talks by well-known scholars, and presentations of submitted papers and posters. It was hosted by the University of Nebraska-Lincoln and took place between the 9th and 12th of July, 2012. The chapters of this volume are based on presentations from the meeting and reflect the latest work in the field. Topics with a primarily measurement focus include studies of item response theory, computerized adaptive testing, cognitive diagnostic modeling, and psychological scaling. Additional psychometric topics relate to structural equation modeling, factor analysis, causal modeling, mediation, missing data methods, and longitudinal data analysis, among others. The papers in this volume will be especially useful for researchers (graduate students and other quantitative researchers) in the social sciences who use quantitative methods, particularly psychologists. Most readers will benefit from some prior knowledge of statistical methods in reading the chapters.

Intelligent Systems can be defined as systems whose design, mainly based on computational techniques, is supported, in some parts, by operations and processing skills inspired by human reasoning and behaviour. Intelligent Systems must typically operate in a scenario in which non-linearities are the rule and not as a disturbing effect to be corrected. Finally, Intelligent Systems also have to incorporate advanced sensory technology in order to simplify man-machine interactions. Several algorithms are currently the ordinary tools of Intelligent Systems. This book contains a selection of contributions regarding Intelligent Systems by experts in diverse fields. Topics discussed in the book are: Applications of Intelligent Systems in Modelling and Prediction of Environmental Changes, Cellular Neural Networks for NonLinear Filtering, NNs for Signal Processing, Image Processing, Transportation Intelligent Systems, Intelligent Techniques in Power Electronics, Applications in Medicine and Surgery, Hardware Implementation and Learning of NNs.

Predictive Modeling of Drug Sensitivity gives an overview of drug sensitivity modeling for personalized medicine that includes data characterizations, modeling techniques, applications, and research challenges. It covers the major mathematical techniques used for modeling drug sensitivity, and includes the requisite biological knowledge to guide a user to apply the mathematical tools in different biological scenarios. This book is an ideal reference for computer scientists, engineers, computational biologists, and mathematicians who want to understand and apply multiple approaches and methods to drug sensitivity modeling. The reader will learn a broad range of mathematical and computational techniques applied to the modeling of drug sensitivity, biological concepts, and measurement techniques crucial to drug sensitivity modeling, how to design a combination of drugs under different constraints, and the applications of drug sensitivity prediction methodologies. - Applies mathematical and computational approaches to biological problems - Covers all aspects of drug sensitivity modeling, starting from initial data generation to final experimental validation - Includes the latest results on drug sensitivity modeling that is based on updated research findings - Provides information on existing data and software resources for applying the mathematical and computational tools available

Covering theory, algorithms, and methodologies, as well as data mining technologies, Data Mining for Bioinformatics provides a comprehensive discussion of data-intensive computations used in data mining with applications in bioinformatics. It supplies a broad, yet in-depth, overview of the application domains of data mining for bioinformatics to help readers from both biology and computer science backgrounds gain an enhanced understanding of this cross-disciplinary field. The book offers authoritative coverage of data mining techniques, technologies, and frameworks used for storing, analyzing, and extracting knowledge from large databases in the bioinformatics domains, including genomics and proteomics. It begins by describing the evolution of bioinformatics and highlighting the challenges that can be addressed using data mining techniques. Introducing the various data mining techniques that can be employed in biological databases, the text is organized into four sections: Supplies a complete overview of the evolution of the field and its intersection with computational learning Describes the role of data mining in analyzing large biological databases—explaining the breath of the various feature selection and feature extraction techniques that data mining has to offer Focuses on concepts of unsupervised learning using clustering techniques and its application to large biological data Covers supervised learning using classification techniques most commonly used in bioinformatics—addressing the need for validation and benchmarking of inferences derived using either clustering or classification The book describes the various biological databases prominently referred to in bioinformatics and includes a detailed list of the applications of advanced clustering algorithms used in bioinformatics. Highlighting the challenges encountered during the application of classification on biological databases, it considers systems of both single and ensemble classifiers and shares effort-saving tips for model selection and performance estimation strategies.

The two volume set LNCS 3696 and LNCS 3697 constitutes the refereed proceedings of the 15th International Conference on Artificial Neural Networks, ICANN 2005, held in Warsaw, Poland in September 2005. The over 600 papers submitted to ICANN 2005 were thoroughly reviewed and carefully selected for presentation. The first volume includes 106 contributions related to Biological Inspirations; topics addressed are modeling the brain and cognitive functions, development of cognitive powers in embodied systems spiking neural networks, associative memory models, models of biological functions, projects in the area of neuroIT, evolutionary and other biological inspirations, self-organizing maps and their applications, computer vision, face recognition and detection, sound and speech recognition, bioinformatics, biomedical applications, and information- theoretic concepts in biomedical data analysis. The second volume contains 162 contributions related to Formal Models and their Applications and deals with new neural network models, supervised learning algorithms, ensemble-based learning, unsupervised learning, recurent neural networks, reinforcement learning, bayesian approaches to learning, learning theory, artificial neural networks for system modeling, decision making, optimalization and control, knowledge extraction from neural networks, temporal data analysis, prediction and forecasting, support vector machines and kernel-based methods, soft computing methods for data representation, analysis and processing, data fusion for industrial, medical and environmental applications, non-linear predictive models for speech processing, intelligent multimedia and semantics, applications to natural language processing, various applications, computational intelligence in games, and issues in hardware implementation.