Author: Grace Wahba
Publisher:
ISBN:
Category :
Languages : en
Pages : 45
Book Description
The relationship between certain regularization methods for solving ill posed linear operator equations and ridge methods in regression problems is described. The regularization estimates we describe may be viewed as ridge estimates in a (reproducing kernel) Hilbert space H. When the solution is known a priori to be in some closed, convex set in H, for example, the set of nonnegative functions, or the set of monotone functions, then one can propose regularized estimates subject to side conditions such as nonnegativity, monotonicity, etc. Some applications in medicine and meteorology are described. We describe the method of generalized cross validation for choosing the smoothing (or ridge) parameter in the presence of a family of linear inequality constraints. Some successful numerical examples, solving ill posed convolution equations with noisy data, subject to nonnegativity constraints, are presented. The technique appears to be quite successful in adding information, doing nearly the optimal amount of smoothing, and resolving distinct peaks in the solution which have been blurred by the convolution operation. (Author).
Constrained Regularization for Ill Posed Linear Operator Equations, with Applications in Meteorology and Medicine
Author: Grace Wahba
Publisher:
ISBN:
Category :
Languages : en
Pages : 45
Book Description
The relationship between certain regularization methods for solving ill posed linear operator equations and ridge methods in regression problems is described. The regularization estimates we describe may be viewed as ridge estimates in a (reproducing kernel) Hilbert space H. When the solution is known a priori to be in some closed, convex set in H, for example, the set of nonnegative functions, or the set of monotone functions, then one can propose regularized estimates subject to side conditions such as nonnegativity, monotonicity, etc. Some applications in medicine and meteorology are described. We describe the method of generalized cross validation for choosing the smoothing (or ridge) parameter in the presence of a family of linear inequality constraints. Some successful numerical examples, solving ill posed convolution equations with noisy data, subject to nonnegativity constraints, are presented. The technique appears to be quite successful in adding information, doing nearly the optimal amount of smoothing, and resolving distinct peaks in the solution which have been blurred by the convolution operation. (Author).
Publisher:
ISBN:
Category :
Languages : en
Pages : 45
Book Description
The relationship between certain regularization methods for solving ill posed linear operator equations and ridge methods in regression problems is described. The regularization estimates we describe may be viewed as ridge estimates in a (reproducing kernel) Hilbert space H. When the solution is known a priori to be in some closed, convex set in H, for example, the set of nonnegative functions, or the set of monotone functions, then one can propose regularized estimates subject to side conditions such as nonnegativity, monotonicity, etc. Some applications in medicine and meteorology are described. We describe the method of generalized cross validation for choosing the smoothing (or ridge) parameter in the presence of a family of linear inequality constraints. Some successful numerical examples, solving ill posed convolution equations with noisy data, subject to nonnegativity constraints, are presented. The technique appears to be quite successful in adding information, doing nearly the optimal amount of smoothing, and resolving distinct peaks in the solution which have been blurred by the convolution operation. (Author).
Inverse and Ill-Posed Problems
Author: Heinz W. Engl
Publisher: Elsevier
ISBN: 1483272656
Category : Mathematics
Languages : en
Pages : 585
Book Description
Inverse and Ill-Posed Problems is a collection of papers presented at a seminar of the same title held in Austria in June 1986. The papers discuss inverse problems in various disciplines; mathematical solutions of integral equations of the first kind; general considerations for ill-posed problems; and the various regularization methods for integral and operator equations of the first kind. Other papers deal with applications in tomography, inverse scattering, detection of radiation sources, optics, partial differential equations, and parameter estimation problems. One paper discusses three topics on ill-posed problems, namely, the imposition of specified types of discontinuities on solutions of ill-posed problems, the use of generalized cross validation as a data based termination rule for iterative methods, and also a parameter estimation problem in reservoir modeling. Another paper investigates a statistical method to determine the truncation level in Eigen function expansions and for Fredholm equations of the first kind where the data contains some errors. Another paper examines the use of singular function expansions in the inversion of severely ill-posed problems arising in confocal scanning microscopy, particle sizing, and velocimetry. The collection can benefit many mathematicians, students, and professor of calculus, statistics, and advanced mathematics.
Publisher: Elsevier
ISBN: 1483272656
Category : Mathematics
Languages : en
Pages : 585
Book Description
Inverse and Ill-Posed Problems is a collection of papers presented at a seminar of the same title held in Austria in June 1986. The papers discuss inverse problems in various disciplines; mathematical solutions of integral equations of the first kind; general considerations for ill-posed problems; and the various regularization methods for integral and operator equations of the first kind. Other papers deal with applications in tomography, inverse scattering, detection of radiation sources, optics, partial differential equations, and parameter estimation problems. One paper discusses three topics on ill-posed problems, namely, the imposition of specified types of discontinuities on solutions of ill-posed problems, the use of generalized cross validation as a data based termination rule for iterative methods, and also a parameter estimation problem in reservoir modeling. Another paper investigates a statistical method to determine the truncation level in Eigen function expansions and for Fredholm equations of the first kind where the data contains some errors. Another paper examines the use of singular function expansions in the inversion of severely ill-posed problems arising in confocal scanning microscopy, particle sizing, and velocimetry. The collection can benefit many mathematicians, students, and professor of calculus, statistics, and advanced mathematics.
Function Estimates
Author: James Stephen Marron
Publisher: American Mathematical Soc.
ISBN: 0821850628
Category : Mathematics
Languages : en
Pages : 190
Book Description
This volume collects together papers presented at the 1985 Conference in Function Estimation held at Humboldt State University. The papers focus especially on various types of spline estimations and convolution problems. The use of estimation and approximation methods as applied to geophysics, numerical analysis, and nonparametric statistics was a special feature of this conference.
Publisher: American Mathematical Soc.
ISBN: 0821850628
Category : Mathematics
Languages : en
Pages : 190
Book Description
This volume collects together papers presented at the 1985 Conference in Function Estimation held at Humboldt State University. The papers focus especially on various types of spline estimations and convolution problems. The use of estimation and approximation methods as applied to geophysics, numerical analysis, and nonparametric statistics was a special feature of this conference.
Spline Models for Observational Data
Author: Grace Wahba
Publisher: SIAM
ISBN: 9781611970128
Category : Mathematics
Languages : en
Pages : 181
Book Description
This book serves well as an introduction into the more theoretical aspects of the use of spline models. It develops a theory and practice for the estimation of functions from noisy data on functionals. The simplest example is the estimation of a smooth curve, given noisy observations on a finite number of its values. The estimate is a polynomial smoothing spline. By placing this smoothing problem in the setting of reproducing kernel Hilbert spaces, a theory is developed which includes univariate smoothing splines, thin plate splines in d dimensions, splines on the sphere, additive splines, and interaction splines in a single framework. A straightforward generalization allows the theory to encompass the very important area of (Tikhonov) regularization methods for ill-posed inverse problems. Convergence properties, data based smoothing parameter selection, confidence intervals, and numerical methods are established which are appropriate to a wide variety of problems which fall within this framework. Methods for including side conditions and other prior information in solving ill-posed inverse problems are included. Data which involves samples of random variables with Gaussian, Poisson, binomial, and other distributions are treated in a unified optimization context. Experimental design questions, i.e., which functionals should be observed, are studied in a general context. Extensions to distributed parameter system identification problems are made by considering implicitly defined functionals.
Publisher: SIAM
ISBN: 9781611970128
Category : Mathematics
Languages : en
Pages : 181
Book Description
This book serves well as an introduction into the more theoretical aspects of the use of spline models. It develops a theory and practice for the estimation of functions from noisy data on functionals. The simplest example is the estimation of a smooth curve, given noisy observations on a finite number of its values. The estimate is a polynomial smoothing spline. By placing this smoothing problem in the setting of reproducing kernel Hilbert spaces, a theory is developed which includes univariate smoothing splines, thin plate splines in d dimensions, splines on the sphere, additive splines, and interaction splines in a single framework. A straightforward generalization allows the theory to encompass the very important area of (Tikhonov) regularization methods for ill-posed inverse problems. Convergence properties, data based smoothing parameter selection, confidence intervals, and numerical methods are established which are appropriate to a wide variety of problems which fall within this framework. Methods for including side conditions and other prior information in solving ill-posed inverse problems are included. Data which involves samples of random variables with Gaussian, Poisson, binomial, and other distributions are treated in a unified optimization context. Experimental design questions, i.e., which functionals should be observed, are studied in a general context. Extensions to distributed parameter system identification problems are made by considering implicitly defined functionals.
Advances in Large Margin Classifiers
Author: Alexander J. Smola
Publisher: MIT Press
ISBN: 9780262194488
Category : Computers
Languages : en
Pages : 436
Book Description
The book provides an overview of recent developments in large margin classifiers, examines connections with other methods (e.g., Bayesian inference), and identifies strengths and weaknesses of the method, as well as directions for future research. The concept of large margins is a unifying principle for the analysis of many different approaches to the classification of data from examples, including boosting, mathematical programming, neural networks, and support vector machines. The fact that it is the margin, or confidence level, of a classification--that is, a scale parameter--rather than a raw training error that matters has become a key tool for dealing with classifiers. This book shows how this idea applies to both the theoretical analysis and the design of algorithms. The book provides an overview of recent developments in large margin classifiers, examines connections with other methods (e.g., Bayesian inference), and identifies strengths and weaknesses of the method, as well as directions for future research. Among the contributors are Manfred Opper, Vladimir Vapnik, and Grace Wahba.
Publisher: MIT Press
ISBN: 9780262194488
Category : Computers
Languages : en
Pages : 436
Book Description
The book provides an overview of recent developments in large margin classifiers, examines connections with other methods (e.g., Bayesian inference), and identifies strengths and weaknesses of the method, as well as directions for future research. The concept of large margins is a unifying principle for the analysis of many different approaches to the classification of data from examples, including boosting, mathematical programming, neural networks, and support vector machines. The fact that it is the margin, or confidence level, of a classification--that is, a scale parameter--rather than a raw training error that matters has become a key tool for dealing with classifiers. This book shows how this idea applies to both the theoretical analysis and the design of algorithms. The book provides an overview of recent developments in large margin classifiers, examines connections with other methods (e.g., Bayesian inference), and identifies strengths and weaknesses of the method, as well as directions for future research. Among the contributors are Manfred Opper, Vladimir Vapnik, and Grace Wahba.
Advances in Kernel Methods
Author: Bernhard Schölkopf
Publisher: MIT Press
ISBN: 9780262194167
Category : Computers
Languages : en
Pages : 400
Book Description
A young girl hears the story of her great-great-great-great- grandfather and his brother who came to the United States to make a better life for themselves helping to build the transcontinental railroad.
Publisher: MIT Press
ISBN: 9780262194167
Category : Computers
Languages : en
Pages : 400
Book Description
A young girl hears the story of her great-great-great-great- grandfather and his brother who came to the United States to make a better life for themselves helping to build the transcontinental railroad.
The Journal of Integral Equations and Applications
Author:
Publisher:
ISBN:
Category : Integral equations
Languages : en
Pages : 636
Book Description
Publisher:
ISBN:
Category : Integral equations
Languages : en
Pages : 636
Book Description
Technical Report
Author: University of Wisconsin--Madison. Department of Statistics
Publisher:
ISBN:
Category :
Languages : en
Pages : 488
Book Description
Publisher:
ISBN:
Category :
Languages : en
Pages : 488
Book Description
Machine Learning Techniques for Space Weather
Author: Enrico Camporeale
Publisher: Elsevier
ISBN: 0128117893
Category : Science
Languages : en
Pages : 454
Book Description
Machine Learning Techniques for Space Weather provides a thorough and accessible presentation of machine learning techniques that can be employed by space weather professionals. Additionally, it presents an overview of real-world applications in space science to the machine learning community, offering a bridge between the fields. As this volume demonstrates, real advances in space weather can be gained using nontraditional approaches that take into account nonlinear and complex dynamics, including information theory, nonlinear auto-regression models, neural networks and clustering algorithms. Offering practical techniques for translating the huge amount of information hidden in data into useful knowledge that allows for better prediction, this book is a unique and important resource for space physicists, space weather professionals and computer scientists in related fields. Collects many representative non-traditional approaches to space weather into a single volume Covers, in an accessible way, the mathematical background that is not often explained in detail for space scientists Includes free software in the form of simple MATLAB® scripts that allow for replication of results in the book, also familiarizing readers with algorithms
Publisher: Elsevier
ISBN: 0128117893
Category : Science
Languages : en
Pages : 454
Book Description
Machine Learning Techniques for Space Weather provides a thorough and accessible presentation of machine learning techniques that can be employed by space weather professionals. Additionally, it presents an overview of real-world applications in space science to the machine learning community, offering a bridge between the fields. As this volume demonstrates, real advances in space weather can be gained using nontraditional approaches that take into account nonlinear and complex dynamics, including information theory, nonlinear auto-regression models, neural networks and clustering algorithms. Offering practical techniques for translating the huge amount of information hidden in data into useful knowledge that allows for better prediction, this book is a unique and important resource for space physicists, space weather professionals and computer scientists in related fields. Collects many representative non-traditional approaches to space weather into a single volume Covers, in an accessible way, the mathematical background that is not often explained in detail for space scientists Includes free software in the form of simple MATLAB® scripts that allow for replication of results in the book, also familiarizing readers with algorithms
Methods and Procedures for the Verification and Validation of Artificial Neural Networks
Author: Brian J. Taylor
Publisher: Springer Science & Business Media
ISBN: 0387294856
Category : Computers
Languages : en
Pages : 280
Book Description
Neural networks are members of a class of software that have the potential to enable intelligent computational systems capable of simulating characteristics of biological thinking and learning. Currently no standards exist to verify and validate neural network-based systems. NASA Independent Verification and Validation Facility has contracted the Institute for Scientific Research, Inc. to perform research on this topic and develop a comprehensive guide to performing V&V on adaptive systems, with emphasis on neural networks used in safety-critical or mission-critical applications. Methods and Procedures for the Verification and Validation of Artificial Neural Networks is the culmination of the first steps in that research. This volume introduces some of the more promising methods and techniques used for the verification and validation (V&V) of neural networks and adaptive systems. A comprehensive guide to performing V&V on neural network systems, aligned with the IEEE Standard for Software Verification and Validation, will follow this book.
Publisher: Springer Science & Business Media
ISBN: 0387294856
Category : Computers
Languages : en
Pages : 280
Book Description
Neural networks are members of a class of software that have the potential to enable intelligent computational systems capable of simulating characteristics of biological thinking and learning. Currently no standards exist to verify and validate neural network-based systems. NASA Independent Verification and Validation Facility has contracted the Institute for Scientific Research, Inc. to perform research on this topic and develop a comprehensive guide to performing V&V on adaptive systems, with emphasis on neural networks used in safety-critical or mission-critical applications. Methods and Procedures for the Verification and Validation of Artificial Neural Networks is the culmination of the first steps in that research. This volume introduces some of the more promising methods and techniques used for the verification and validation (V&V) of neural networks and adaptive systems. A comprehensive guide to performing V&V on neural network systems, aligned with the IEEE Standard for Software Verification and Validation, will follow this book.