Meta-Algorithmics : Patterns for Robust, Low Cost, High Quality Systems.
Title:
Meta-Algorithmics : Patterns for Robust, Low Cost, High Quality Systems.
Author:
Simske, Steven J.
ISBN:
9781118626702
Personal Author:
Edition:
1st ed.
Physical Description:
1 online resource (388 pages)
Series:
Wiley - IEEE
Contents:
META-ALGORITHMICS -- Contents -- Acknowledgments -- 1 Introduction and Overview -- 1.1 Introduction -- 1.2 Why Is This Book Important? -- 1.3 Organization of the Book -- 1.4 Informatics -- 1.5 Ensemble Learning -- 1.6 Machine Learning/Intelligence -- 1.6.1 Regression and Entropy -- 1.6.2 SVMs and Kernels -- 1.6.3 Probability -- 1.6.4 Unsupervised Learning -- 1.6.5 Dimensionality Reduction -- 1.6.6 Optimization and Search -- 1.7 Artificial Intelligence -- 1.7.1 Neural Networks -- 1.7.2 Genetic Algorithms -- 1.7.3 Markov Models -- 1.8 Data Mining/Knowledge Discovery -- 1.9 Classification -- 1.10 Recognition -- 1.11 System-Based Analysis -- 1.12 Summary -- References -- 2 Parallel Forms of Parallelism -- 2.1 Introduction -- 2.2 Parallelism by Task -- 2.2.1 Definition -- 2.2.2 Application to Algorithms and Architectures -- 2.2.3 Application to Scheduling -- 2.3 Parallelism by Component -- 2.3.1 Definition and Extension to Parallel-Conditional Processing -- 2.3.2 Application to Data Mining, Search, and Other Algorithms -- 2.3.3 Application to Software Development -- 2.4 Parallelism by Meta-algorithm -- 2.4.1 Meta-algorithmics and Algorithms -- 2.4.2 Meta-algorithmics and Systems -- 2.4.3 Meta-algorithmics and Parallel Processing -- 2.4.4 Meta-algorithmics and Data Collection -- 2.4.5 Meta-algorithmics and Software Development -- 2.5 Summary -- References -- 3 Domain Areas: Where Are These Relevant? -- 3.1 Introduction -- 3.2 Overview of the Domains -- 3.3 Primary Domains -- 3.3.1 Document Understanding -- 3.3.2 Image Understanding -- 3.3.3 Biometrics -- 3.3.4 Security Printing -- 3.4 Secondary Domains -- 3.4.1 Image Segmentation -- 3.4.2 Speech Recognition -- 3.4.3 Medical Signal Processing -- 3.4.4 Medical Imaging -- 3.4.5 Natural Language Processing -- 3.4.6 Surveillance -- 3.4.7 Optical Character Recognition -- 3.4.8 Security Analytics.
3.5 Summary -- References -- 4 Applications of Parallelism by Task -- 4.1 Introduction -- 4.2 Primary Domains -- 4.2.1 Document Understanding -- 4.2.2 Image Understanding -- 4.2.3 Biometrics -- 4.2.4 Security Printing -- 4.3 Summary -- References -- 5 Application of Parallelism by Component -- 5.1 Introduction -- 5.2 Primary Domains -- 5.2.1 Document Understanding -- 5.2.2 Image Understanding -- 5.2.3 Biometrics -- 5.2.4 Security Printing -- 5.3 Summary -- References -- 6 Introduction to Meta-algorithmics -- 6.1 Introduction -- 6.2 First-Order Meta-algorithmics -- 6.2.1 Sequential Try -- 6.2.2 Constrained Substitute -- 6.2.3 Voting and Weighted Voting -- 6.2.4 Predictive Selection -- 6.2.5 Tessellation and Recombination -- 6.3 Second-Order Meta-algorithmics -- 6.3.1 Confusion Matrix and Weighted Confusion Matrix -- 6.3.2 Confusion Matrix with Output Space Transformation (Probability Space Transformation) -- 6.3.3 Tessellation and Recombination with Expert Decisioner -- 6.3.4 Predictive Selection with Secondary Engines -- 6.3.5 Single Engine with Required Precision -- 6.3.6 Majority Voting or Weighted Confusion Matrix -- 6.3.7 Majority Voting or Best Engine -- 6.3.8 Best Engine with Differential Confidence or Second Best Engine -- 6.3.9 Best Engine with Absolute Confidence or Weighted Confusion Matrix -- 6.4 Third-Order Meta-algorithmics -- 6.4.1 Feedback -- 6.4.2 Proof by Task Completion -- 6.4.3 Confusion Matrix for Feedback -- 6.4.4 Expert Feedback -- 6.4.5 Sensitivity Analysis -- 6.4.6 Regional Optimization (Extended Predictive Selection) -- 6.4.7 Generalized Hybridization -- 6.5 Summary -- References -- 7 First-Order Meta-algorithmics and Their Applications -- 7.1 Introduction -- 7.2 First-Order Meta-algorithmics and the "Black Box" -- 7.3 Primary Domains -- 7.3.1 Document Understanding -- 7.3.2 Image Understanding -- 7.3.3 Biometrics.
7.3.4 Security Printing -- 7.4 Secondary Domains -- 7.4.1 Medical Signal Processing -- 7.4.2 Medical Imaging -- 7.4.3 Natural Language Processing -- 7.5 Summary -- References -- 8 Second-Order Meta-algorithmics and Their Applications -- 8.1 Introduction -- 8.2 Second-Order Meta-algorithmics and Targeting the "Fringes" -- 8.3 Primary Domains -- 8.3.1 Document Understanding -- 8.3.2 Image Understanding -- 8.3.3 Biometrics -- 8.3.4 Security Printing -- 8.4 Secondary Domains -- 8.4.1 Image Segmentation -- 8.4.2 Speech Recognition -- 8.5 Summary -- References -- 9 Third-Order Meta-algorithmics and Their Applications -- 9.1 Introduction -- 9.2 Third-Order Meta-algorithmic Patterns -- 9.2.1 Examples Covered -- 9.2.2 Training-Gap-Targeted Feedback -- 9.3 Primary Domains -- 9.3.1 Document Understanding -- 9.3.2 Image Understanding -- 9.3.3 Biometrics -- 9.3.4 Security Printing -- 9.4 Secondary Domains -- 9.4.1 Surveillance -- 9.4.2 Optical Character Recognition -- 9.4.3 Security Analytics -- 9.5 Summary -- References -- 10 Building More Robust Systems -- 10.1 Introduction -- 10.2 Summarization -- 10.2.1 Ground Truthing for Meta-algorithmics -- 10.2.2 Meta-algorithmics for Keyword Generation -- 10.3 Cloud Systems -- 10.4 Mobile Systems -- 10.5 Scheduling -- 10.6 Classification -- 10.7 Summary -- Reference -- 11 The Future -- 11.1 Recapitulation -- 11.2 The Pattern of All Patience -- 11.3 Beyond the Pale -- 11.4 Coming Soon -- 11.5 Summary -- References -- Index.
Abstract:
The confluence of cloud computing, parallelism and advanced machine intelligence approaches has created a world in which the optimum knowledge system will usually be architected from the combination of two or more knowledge-generating systems. There is a need, then, to provide a reusable, broadly-applicable set of design patterns to empower the intelligent system architect to take advantage of this opportunity. This book explains how to design and build intelligent systems that are optimized for changing system requirements (adaptability), optimized for changing system input (robustness), and optimized for one or more other important system parameters (e.g., accuracy, efficiency, cost). It provides an overview of traditional parallel processing which is shown to consist primarily of task and component parallelism; before introducing meta-algorithmic parallelism which is based on combining two or more algorithms, classification engines or other systems. Key features: Explains the entire roadmap for the design, testing, development, refinement, deployment and statistics-driven optimization of building systems for intelligence Offers an accessible yet thorough overview of machine intelligence, in addition to having a strong image processing focus Contains design patterns for parallelism, especially meta-algorithmic parallelism - simply conveyed, reusable and proven effective that can be readily included in the toolbox of experts in analytics, system architecture, big data, security and many other science and engineering disciplines Connects algorithms and analytics to parallelism, thereby illustrating a new way of designing intelligent systems compatible with the tremendous changes in the computing world over the past decade Discusses application of the approaches to a wide number of fields; primarily, document understanding, image understanding,
biometrics and security printing Companion website contains sample code and data sets.
Local Note:
Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2017. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries.
Genre:
Electronic Access:
Click to View