Artificial intelligence with uncertainty / Deyi Li and Yi Du.

Format:

Book

Author/Creator:

Li, Deyi, 1944- author.

Du, Yi, 1971- author.

Language:

English

Subjects (All):

Artificial intelligence.

Uncertainty (Information theory).

Physical Description:

xx, 290 pages ; 25 cm

Edition:

Second edition.

Place of Publication:

Boca Raton, FL : CRC Press, [2017]

Summary:

Artificial Intelligence with Uncertainty, Second Edition is concerned with uncertainty in artificial intelligence. Uncertainty in human intelligence and human knowledge is universal, objective, and beautiful but also difficult to simulate. The models and reasoning mechanisms for artificial intelligence with uncertainty are discussed in this book through cloud model, cloud transformation, data field, intelligent control, and swarm intelligence. The second edition of Artificial Intelligence with Uncertainty adds coverage of soft computing, semantic computing, granular computing, cloud computing, the Internet of Things, and big data over the years. It also, Discusses the objectivity, universality, and positivity of the existence of uncertainty in human knowledge and intelligence, Studies the mathematical basis, feature, representation, model, inference mechanism, certainty of the uncertain thinking activities of artificial intelligence with uncertainty, Expands hierarchically from the cloud model used for qualitative and quantitative transformation and the physical method of cognition to data mining, knowledge discovery, and intelligent, control, Identifies the regularity in uncertain knowledge and intelligent processing, and finally, Forecasts the development direction of research on artificial intelligence with uncertainty. The authors' research has given the second edition of Artificial Intelligence with Uncertainty a distinct character, making it valuable for scientists, engineers, and graduate students engaged in cognitive science, Al theory, and knowledge engineering. Book jacket.

Contents:

1 Artificial Intelligence Challenged by Uncertainty 1

1.1 The Uncertainly of Human Intelligence 2

1.1.1 The Charm of Uncertainty 2

1.1.2 The World of Entropy 6

1.2 Sixty Years of Artificial Intelligence Development 8

1.2.1 The Dartmouth Symposium 8

1.2.1.1 Collision between Different Disciplines 8

1.2.1.2 Ups and Downs in Development 9

1.2.2 Goals Evolve over Time 10

1.2.2.1 Turing Test 10

1.2.2.2 Proving Theorems by Machine 11

1.2.2.3 Rivalry between Kasparov and Deep Blue 11

1.2.2.4 Thinking Machine 12

1.2.2.5 Artificial Life 13

1.2.3 Significant Achievements in AI over the Past 60 Years 14

1.3 Research Methods for AI 17

1.3.1 Symbolism 17

1.3.2 Connectionism 20

1.3.3 Behaviorism 21

1.4 Interdisciplinary Trends in AI 23

1.4.1 Brain Science and AI 23

1.4.2 Cognitive Science and AI 26

1.4.3 Network Science and AI 27

1.4.4 Great Breakthroughs to Be Achieved by Interdisciplinary Research 29

References 30

2 Cloud Model: A Cognitive Model for Qualitative and Quantitative Transformation 31

2.1 Starting Points for the Study of Artificial Intelligence with Uncertainty 31

2.1.1 Multiple Starting Points 31

2.1.2 Keeping in Mind Concepts of Natural Languages 32

2.1.3 Randomness and Fuzziness in Concepts 33

2.2 Using Cloud Models to Represent Uncertainties of Concepts 34

2.2.1 Cloud and Cloud Drops 35

2.2.2 The Cloud's Digital Characteristics 36

2.2.3 Various Types of Cloud Models 37

2.3 Algorithm of Forward Gaussian Cloud 39

2.3.1 Description 40

2.3.1.1 Forward Gaussian Cloud Algorithm 40

2.3.1.2 Two-Dimensional Forward Gaussian Algorithm 41

2.3.2 Contributions Made by Cloud Drops to the Concept 42

2.3.3 Using Gaussian Cloud Model to Understand the 24 Solar Terms in the Chinese Lunar Calendar 44

2.4 Mathematical Properties of the Gaussian Cloud 45

2.4.1 Statistical Analysis of the Distribution of Cloud Drops 45

2.4.2 Statistical Analysis of Certainty Degree of Cloud Drops 49

2.4.3 Expectation Curves of Gaussian Cloud 52

2.4.4 From Cloud to Fog 54

2.5 Algorithm of Backward Gaussian Cloud 56

2.5.1 Description 57

2.5.1.1 Backward Gaussian Cloud Algorithm with Certainty Degree 57

2.5.1.2 Backward Cloud Algorithm Based on the First-Order Absolute Central Moment and the Second-Order Central Moment 59

2.5.1.3 Backward Cloud Algorithm Based on the Second- and Fourth-Order Central Moments of Samples 60

2.5.2 Parameter Estimation and Error Analysis of Backward Gaussian Cloud 60

2.5.2.1 Error Analysis of Ex 60

2.5.2.2 Error Analysis of En and He 61

2.5.2.3 Determining the Number of Samples under the Condition of Given Errors and Confidence Level 62

2.6 Further Understanding of the Cloud Model 63

2.6.1 Judgment Shooting 63

2.6.2 Fractal with Uncertainty 68

2.7 Universality of the Gaussian Cloud 71

2.7.1 Universality of Gaussian Distribution 72

2.7.2 Universality of Bell-Shaped Membership Function 73

2.7.3 Universal Relevance of Gaussian Cloud 76

References 78

3 Gaussian Cloud Transformation 79

3.1 Terminology in Granular Computing 80

3.1.1 Scale, Level, and Granularity 80

3.1.2 Concept Tree and Pan-Concept Tree 82

3.2 Gaussian Transformation 83

3.2.1 Parameter Estimation of Gaussian Transform 84

3.2.2 Gaussian Transform Algorithm 86

3.3 Gaussian Cloud Transformation 89

3.3.1 From Gaussian Transformation to Gaussian Cloud Transformation 90

3.3.2 Heuristic Gaussian Cloud Transformation 92

3.3.3 Adaptive Gaussian Cloud Transformation 96

3.3.4 High-Dimensional Gaussian Cloud Transformation 100

3.4 Gaussian Cloud Transformation for Image Segmentation 102

3.4.1 Detection of the Transition Zone in Images 102

3.4.2 Differential Object Extraction in Image 107

References 114

4 Data Fields and Topological Potential 115

4.1 Data Field 115

4.1.1 Using Field to Describe Interactive Objects 115

4.1.2 From Physical Field to Data Field 117

4.1.3 Potential Field and Force Field of Data 121

4.1.4 Selection of Influence Coefficient in Field Function 130

4.2 Clustering Based on Data Field 133

4.2.1 Uncertainty in Classification and Clustering 133

4.2.2 Dynamic Clustering Based on Data Field 134

4.2.2.1 Selecting Representative Objects through Mass Estimation 135

4.2.2.2 Initial Clustering of Data Samples 136

4.2.2.3 Dynamic Clustering of Representative Objects 137

4.2.3 Expression Clustering of Human Face Images Based on Data Field 141

4.2.3.1 Feature Extraction Based on Face Image Data Field 144

4.2.3.2 Recognition of Facial Expression Cluster Based on K-L Transformation and Second-Order Data Field 148

4.3 Complex Network Research Based on Topological Potential 150

4.3.1 From Data Field to Topological Potential 151

4.3.2 Important Network Nodes Detected with Topological Potential 152

4.3.3 Network Community Discovery Based on Topological Potential 158

4.3.4 Hot Entries in Wikipedia Discovered with Topological Potential 163

References 171

5 Reasoning and Control of Qualitative Knowledge 173

5.1 Cloud Reasoning 174

5.1.1 Using a Cloud Model to Construct Qualitative Rules 174

5.1.1.1 One-Dimensional Precondition Cloud Generator 175

5.1.1.2 Postcondition Cloud Generator 175

5.1.1.3 Single-Condition-Single-Rule Generator 176

5.1.2 Generation of Rule Sets 179

5.2 Cloud Control 180

5.2.1 Mechanism of Cloud Control 180

5.2.2 Theoretical Explanation of the Mamdani Fuzzy Control Method 190

5.3 Uncertainty Control in Inverted Pendulum 191

5.3.1 Inverted Pendulum System and Its Control 191

5.3.2 Qualitative Control Mechanisms for Single-Link/Double-Link Inverted Pendulums 192

5.3.3 Cloud Control Strategy for a Triple-Link Inverted Pendulum 194

5.3.3.1 Qualitative Analysis of the Triple-Link Inverted Pendulum System 195

5.3.3.2 The Cloud Controller of the Triple-Link Inverted Pendulum System 198

5.3.4 Balancing Patterns of the Inverted Pendulum 199

5.3.4.1 Balancing Patterns of the Single-Link Inverted Pendulum 202

5.3.4.2 Balancing Patterns of the Double-Link Inverted Pendulum 204

5.3.4.3 Balancing Patterns of the Triple-Link Inverted Pendulum 208

5.4 Uncertainty Control in Intelligent Driving 212

5.4.1 Intelligent Driving of Automobiles 213

5.4.1.1 Integration of the Right of Way Radar Map 213

5.4.1.2 Cloud Control Strategy of Intelligent Vehicle 216

5.4.2 Driving Behavior Simulation Based on Intelligent Automobiles 221

References 223

6 Cognitive Physics for Swarm Intelligence 225

6.1 Interaction: The Important Cause of Swarm Intelligence 225

6.1.1 Swarm Intelligence 226

6.1.2 Emergence as a Form to Represent Group Behavior 228

6.2 Application of Cloud Model and Data Field in Swarm Intelligence 230

6.2.1 Cloud Model to Represent Discrete Individual Behavior 230

6.2.2 Data Field to Describe Interactions between Individuals 231

6.3 Typical Case: "Applause Sounded" 232

6.3.1 Cloud Model to Represent People's Applauding Behavior 232

6.3.1.1 Simplification and Modeling of Individual Behavior 232

6.3.1.2 Simplification and Modeling of the Environment 234

6.3.1.3 Initial Distribution and Presentation of Individual Behavior 234

6.3.2 Data Field to Reflect Mutual Spread of Applause 235

6.3.3 Computing Model for "Applause Sounded" 236

6.3.4 Experimental Platform 239

6.3.5 Diversity Analysis of Emergence 242

6.3.6 Guided Applause Synchronization 245

References 248

7 Great Development of Artificial Intelligence with Uncertainty due to Cloud Computing 251

7.1 An Insight into the Contributions and Limitations of Fuzzy Set from the Perspective of a Cloud Model 251

7.1.1 Paradoxical Argument over Fuzzy Logic 251

7.1.2 Dependence of Fuzziness on Randomness 254

7.1.3 From Fuzzy to Uncertainty Reasoning 256

7.2 From Turing Computing to Cloud Computing 258

7.2.1 Cloud Computing beyond the Turing Machine 260

7.2.2 Cloud Computing and Cloud Model 264

7.2.3 Cloud Model Walking between Gaussian and Power Law Distribution 266

7.3 Big Data Calls for AI with Uncertainties 272

7.3.1 From Database to Big Data 272

7.3.2 Network Interaction and Swarm Intelligence 274

7.4 Prospect of AI with Uncertainty 277

References 280.

Notes:

Includes bibliographical references and index.

ISBN:

9781498776264

1498776264

OCLC:

953981477