Machine Intelligence

ISBN-13: 9780470343968 / Angielski / Twarda / 2011 / 248 str.

Haibo He

Opis

This book will advance the understanding and application of self-adaptive intelligent systems; therefore it will potentially benefit the long-term goal of replicating certain levels of brain-like intelligence in complex and networked engineering systems. It will provide new approaches for adaptive systems within uncertain environments. This will provide an opportunity to evaluate the strengths and weaknesses of the current state-of-the-art of knowledge, give rise to new research directions, and educate future professionals in this domain. Self-adaptive intelligent systems have wide applications from military security systems to civilian daily life. In this book, different application problems, including pattern recognition, classification, image recovery, and sequence learning, will be presented to show the capability of the proposed systems in learning, memory, and prediction. Therefore, this book will also provide potential new solutions to many real-world applications.

Recenzje

"This comprehensive introduction to machine intelligence engineering and self–adaptive systems provides an overview of a variety of processes and technologies for the development of artificial intelligence." (Book News, 1 October 2011)

 

Spis treści

Preface.

Acknowledgments.

Chapter 1. Introduction.

1.1 The Machine Intelligence Research.

1.2 The Two–Fold Objectives: Data–Driven and Biologically–Inspired Approaches.

1.3 How to Read this Book.

1.4 Summary and Further Reading.

References.

Chapter 2. Incremental Learning.

2.1 Introduction.

2.2 Problem Foundation.

2.3 An Adaptive Incremental Learning Framework.

2.4 Design of the Mapping Function.

2.5 Case Study.

2.6 Summary.

Chapter 3. Imbalanced Learning.

3.1 Introduction.

3.2 Nature of the Imbalanced Learning.

3.3 Solutions for Imbalanced Learning.

3.4 Assessment Metrics for Imbalanced Learning.

3.5 Opportunities and Challenges.

3.6 Case Study.

3.7 Summary.

Chapter 4. Ensemble Learning.

4.1 Introduction.

4.2 Hypothesis Diversity.

4.3 Developing Multiple Hypotheses.

4.4 Integrating Multiple Hypotheses.

4.5 Case Study.

4.6 Summary.

Chapter 5. Adaptive Dynamic Programming for Machine Intelligence.

5.1 Introduction.

5.2 Fundamental Objectives: Optimization and Prediction.

5.3 ADP for Machine Intelligence.

5.4 Case Study.

5.5 Summary.

Chapter 6. Associative Learning.

6.1 Introduction.

6.2 Associative Learning Mechanism.

6.3 Associative Learning in Hierarchical Neural Networks.

6.4 Case Study.

6.5 Summary.

Chapter 7. Sequence Learning.

7.1 Introduction.

7.2 Foundations for Sequence Learning.

7.3 Sequence Learning in Hierarchical Neural Structure.

7.4 Level 0: A Modified Hebbian Learning Architecture.

7.5 Level 1 to Level N: Sequence Storage, Prediction and Retrieval.

7.6 Memory Requirement.

7.7 Learning and Anticipation of Multiple Sequences.

7.8 Case Study.

7.9 Summary.

Chapter 8. Hardware Design for Machine Intelligence.

8.1 A Final Comment.

References.

O autorze

Haibo He, PhD, is Assistant Professor in the Department of Electrical, Computer, and Biomedical Engineering at the University of Rhode Island. His primary research interest is computational intelligence and self–adaptive systems, including optimization and prediction, biologically inspired machine intelligence, machine learning and data mining, hardware design (VLSI/FPGA) for machine intelligence, as well as various application fields such as smart grid, sensor networks, and cognitive radio networks.

Blurb

Important insights into the challenges of machine intelligence research

Machine intelligence is?the study of the principles, foundations, and designs of adaptive systems that have the ability to learn, predict, optimize, and make decisions to accomplish goals through interaction with uncertain environments. This book advances a fundamental understanding of self–adaptive intelligent systems, helping readers move toward the?long–term goal of replicating certain levels of brain–like intelligence, while also bringing such a level of intelligence closer to reality across many of today′s complex systems.

Self–Adaptive Systems for Machine Intelligence consists of four major sections:

Section 1 introduces self–adaptive systems for machine intelligence research, identifying the research significance and major differences between traditional computation and brain–like intelligence;

Section 2 presents data–driven approaches for machine intelligence research, emphasizing incremental learning, imbalanced learning, and ensemble learning;

Section 3 focuses on biologically inspired machine intelligence research, with adaptive dynamic programming, associative learning, and sequence learning discussed in detail;

Section 4 offers suggestions about critical hardware design considerations such as power consumption, design density, memory, and speed for potentially building complex and integrated self–adaptive systems into real hardware.

Different application problems such as pattern recognition, data classification, adaptive control, and image recovery are presented to show the capability of the proposed systems in learning, prediction, and optimization. The presented principles, architectures, algorithms, and featured case studies not only offer fresh insights into machine intelligence research, but also provide new techniques and solutions across a wide range of real–world applications. All the issues discussed herein are active research topics in the field, making this a valuable resource for graduate students to motivate their research toward master′s and PhD levels. The book is also intended for academic researchers and professionals in the field of computational intelligence/machine learning, industrial researchers and R&D engineers who are interested in adaptive systems, and undergraduates majoring in science or engineering.