Uncertainty in Artificial Intelligence (eBook)

Proceedings of the Ninth Conference on Uncertainty in Artificial Intelligence, The Catholic University of America, Washington, D.C. 1993

David Heckerman, Abe Mamdani (Herausgeber)

eBook Download: PDF

2014 | 1. Auflage
552 Seiten
Elsevier Science (Verlag)
978-1-4832-1451-1 (ISBN)

Infos

Uncertainty in Artificial Intelligence contains the proceedings of the Ninth Conference on Uncertainty in Artificial Intelligence held at the Catholic University of America in Washington, DC, on July 9-11, 1993. The papers focus on methods of reasoning and decision making under uncertainty as applied to problems in artificial intelligence (AI) and cover topics ranging from knowledge acquisition and automated model construction to learning, planning, temporal reasoning, and machine vision. Comprised of 66 chapters, this book begins with a discussion on causality in Bayesian belief networks before turning to a decision theoretic account of conditional ought statements that rectifies glaring deficiencies in classical deontic logic and forms a sound basis for qualitative decision theory. Subsequent chapters explore trade-offs in constructing and evaluating temporal influence diagrams; normative engineering risk management systems; additive belief-network models; and sensitivity analysis for probability assessments in Bayesian networks. Automated model construction and learning as well as algorithms for inference and decision making are also considered. This monograph will be of interest to both students and practitioners in the fields of AI and computer science.

Inhalt

Front Cover 1
Uncertainty in 
 
4 
Copyright Page 5
Table of Contents 6
Preface 10
Acknowledgements 11
Part 1: Foundations 12
Chapter 1. Causality in Bayesian Belief Networks 14
Abstract 14
1 INTRODUCTION 14
2 SIMULTANEOUS EQUATIONS MODELS 16
3 CAUSALITY IN BAYESIAN BELIEF NETWORKS 19
4 CONCLUSION 21
Acknowledgments 22
References 22
Chapter 2. From Conditional Oughts to Qualitative Decision Theory 23
Abstract 23
1 INTRODUCTION 23
2 INFINITESIMAL 
 
 
24 
3 SUMMARY OF RESULTS 25
4 FROM UTILITIES AND BELIEFS TO GOALS AND ACTIONS 25
5 COMBINING ACTIONS AND OBSERVATIONS 27
6 RELATIONS TO OTHER ACCOUNTS 29
7 CONCLUSION 31
Acknowledgements 31
References 31
Part 2: Applications and Empirical Comparisons 32
Chapter 3. A Probabilistic Algorithm for Calculating Structure:Borrowing from Simulated Annealing 34
Abstract 34
1 MOLECULAR STRUCTURE 34
2 THE DATA REPRESENTATION 35
3 EXPERIMENTS PERFORMED 37
4 RESULTS 38
5 DISCUSSION 39
6 CONCLUSIONS 41
Acknowledgements 42
References 42
Chapter 4. A Study of Scaling Issues in Bayesian Belief Networks for Ship Classification 43
Abstract 43
1 Introduction 43
2 Overview 43
3 Network Structure 45
4 Integration of Belief Values 46
5 Discussion 47
6 Conclusion 48
References 48
CHAPTER 5. TRADEOFFS IN CONSTRUCTING AND EVALUATING TEMPORAL INFLUENCE DIAGRAMS 51
Abstract 51
1 INTRODUCTION 51
2 TEMPORAL BAYESIAN NETWORKS 52
3 TID CONSTRUCTION FROM KNOWLEDGE BASES 53
4 DOMAIN-SPECIFIC TIME-SERIES MODELS 54
5 MODEL SELECTION APPROACHES 55
6 EVALUATING TRADEOFFS 57
7 RELATED LITERATURE 57
8 CONCLUSIONS 58
Acknowledgements 58
References 58
Chapter 6. End-User Construction of Influence Diagrams for Bayesian Statistics 59
Abstract 59
1 INTRODUCTION 59
2 STATISTICAL MODEL 60
3 SEMANTIC INTERFACE: THE PATIENT-FLOW DIAGRAM 61
4 METADATA-STATE DIAGRAM: THE COHORT-STATE DIAGRAM 62
5 CONSTRUCTION STEPS 62
6 IMPLEMENTATION 63
7 OTHER WORK 64
8 CONCLUSION 64
Acknowledgments 65
References 65
Chapter 7. On Considering Uncertainty and Alternatives in Low-Level Vision 66
Abstract 66
1 INTRODUCTION 66
2 REGIONS, SEGMENTS, AND SEGMENTATIONS 68
3 SEGMENT-LEVEL UNCERTAINTY 68
4 SEGMENTATION-LEVEL UNCERTAINTY 69
5 REGION-LEVEL UNCERTAINTY 71
6 OBTAINING PRIORS 72
7 ALGORITHMS 72
8 AN EXPERIMENTAL EXAMPLE 73
9 CONCLUSION 73
Acknowledgement 73
References 73
Chapter 8. Forecasting Sleep Apnea with Dynamic Network Models 75
Abstract 75
1 INTRODUCTION 75
2 RELATED WORK 76
3 THE DYNAMIC NETWORK MODEL 76
4 THE DYNEMO IMPLEMENTATION 77
5 THE SLEEP-APNEA FORECASTING PROBLEM 79
6 CONCLUSIONS 81
Acknowledgments 81
References 81
Chapter 9. Normative Engineering Risk Management Systems 83
Abstract 83
1 INTRODUCTION 83
2 ENGINEERING RISK MANAGEMENT SYSTEMS 83
3 ADVANCED RISK MANAGEMENT SYSTEM PROJECT 84
4 NORMATIVE SYSTEM OVERVIEW 85
5 NORMATIVE SYSTEM ACTIVITIES 86
6 RESEARCH ISSUES 89
7 CONCLUSIONS 89
Acknowledgements 89
References 89
Chapter 10. Diagnosis of Multiple Faults: A Sensitivity Analysis 91
Abstract 91
1 INTRODUCTION 91
2 THE MODELS 92
3 EXPERIMENTAL DESIGN 93
4 RESULTS AND DISCUSSION 94
5 Acknowledgment 94
References 94
Part 3: Knowledge Acquisition, Modelling, and Explanation 100
Chapter 11. Additive Belief-Network Models 102
Abstract 102
1 INTRODUCTION 102
2 ADDITIVE MODEL 102
3 SIGNIFICANCE OF ADDITIVEDECOMPOSITION 103
4 FITTING ADDITIVEBELIEF-NETWORK MODELS 105
5 INFERENCE ALGORITHM 107
6 IMPLEMENTATION RESULTS 108
7 CONCLUSIONS 108
Acknowledgments 108
References 108
Chapter 12. Parameter adjustment in Bayes networks.The generalized noisy OR-gate 110
Abstract 110
1 INTRODUCTION 110
2 PARAMETER ADJUSTMENT 111
3 THE GENERALIZED NOISYOR-GATE 113
4 CONCLUSIONS 116
Acknowledgments 116
References 116
Chapter 13. A fuzzy relation-based extension of Reggiavs relational model fordiagnosis handling uncertain and incomplete information 117
Abstract 117
1 INTRODUCTION 117
2 RELATIONAL APPROACH : THECOMPLETELY INFORMED CASE 117
3 THE CASE OF INCOMPLETEINFORMATION 118
4 LINK WITH REGGIA ET AL.'SAPPROACH 119
5 GRADED UNCERTAINTY VS.GRADED INTENSITY OFPRESENCE 120
6 A NEW MODEL BASED ONTWOFOLD FUZZY SETS 121
7 CONCLUDING REMARKS 123
References 124
CHapter 14. Dialectic reasoning with inconsistent information 125
Abstract 125
1 INTRODUCTION 125
2 ARGUMENTATION 126
3 CONSTRUCTING ARGUMENTS 128
4 ACCEPTABILITY CLASSES 128
5 LINGUISTIC QUALIFIERS 129
6 USING PRIORITIES 130
7 Final remarks 131
Acknowledgement 131
References 131
Chapter 15. Causal Independence for Knowledge Acquisition and Inference 133
Abstract 133
1 INTRODUCTION 133
2 A TEMPORAL DEFINITION OFCAUSAL INDEPENDENCE 134
3 A BELIEF-NETWORKREPRESENTATION OF CAUSALINDEPENDENCE 134
4 A REAL-WORLD EXAMPLE 135
5 IMPROVEMENTS IN THEREPRESENTATION FORINFERENCE 136
6 A LIMITATION OF THEREPRESENTATION FORINFERENCE 137
7 AN OBSERVATION ABOUTGENERALITY 137
References 137
Chapter 16. Utility-Based Abstraction and Categorization 139
Abstract 139
1 INTRODUCTION 139
2 ACTIONS UNDERUNCERTAINTY 139
3 ABSTRACTION BYUTILITY-BASED SIMILARITY 140
4 POLYNOMIAL COMPUTATIONOF ABSTRACTIONS 141
5 EXAMPLES OF UTILITY-BASED 
142 
6 DECISIONS WITHABSTRACTIONS 144
7 SUMMARY AND CONCLUSIONS 145
References 146
Chapter 17. Sensitivity Analysis forProbability Assessments in Bayesian Networks 147
Abstract 147
1 INTRODUCTION 147
2 COMPUTING SENSITIVITY VALUES 148
3. EXAMPLE 150
4. INCORPORATING DIRECTESTIMATES OF TARGETDISTRIBUTIONS 151
5. DISCUSSION 152
References 152
APPENDIX: PROOFS OF RESULTS 153
Chapter 18. Causal Modeling 154
Abstract 154
1 INTRODUCTION 154
2 INFORMAL RELATION TO D-GRAPHS 154
3 CAUSAL MODELS 155
4 CONCLUSION 162
5 FURTHER RESEARCH 162
References 162
Chapter 19. Some Complexity Considerations in the Combination of BeliefNetworks 163
Abstract 163
1 INTRODUCTION 163
2 THE GENERAL APPROACH 163
3 PREVIOUS WORK 164
4 FURTHER THEORETICALDEVELOPMENT 165
5 COMPLEXITY ANALYSIS 167
6 SUMMARY 169
References 169
Chapter 20. Deriving a Minimal /-map of a Belief Network Relative to aTarget Ordering of its Nodes 170
Abstract 170
1 INTRODUCTION ANDMOTIVATION 170
2 ILLUSTRATIONS 171
3 PRELIMINARIES 172
4 DERIVING DAG'S 173
5 PROOF OF CORRECTNESS 174
6 SUMMARY 176
Acknowledgement 176
References 176
Chapter 21. Probabilistic Conceptual Network:A Belief Representation Scheme for Utility-Based Categorization 177
1 Introduction 177
2 Integrating Uncertainty and Categorical Knowledge 178
3 An Application in Automated Machining 178
4 Probabilistic Conceptual Network 179
5 Model Construction 182
6 Related Work 182
7 Conclusion 183
Acknowledgements 184
Reference 184
Chapter 22. Reasoning about the Value of Decision-Model Refinement: Methods and Application 185
Abstract 185
1 Introduction 185
2 Expected Values of Decision-Model Refinement 186
3 Control of Refinement 191
4 Discussion and Related Work 192
5 Summary and Conclusions 193
Reference 193
Chapter 23. Mixtures of Gaussians and Minimum Relative Entropy Techniques for Modeling Continuous Uncertainties 194
Abstract 194
1 INTRODUCTION 194
2 INFERENCE AND DECISION MAKING 
 
195 
3 TRANSFORMATIONS TOWARD A DESIRED DISTRIBUTION 197
4 FITTING MIXTURE DISTRIBUTIONS WITH THE EM ALGORITHM 198
5 SELECTING THE MIXTURE SIZE 199
6 CONCLUDING REMARKS 200
References 201
Chapter 24. Valuation Networks and Conditional Independence 202
Abstract 202
1 INTRODUCTION 202
2 VBSs AND CONDITIONAL INDEPENDENCE 203
3 VALUATION NETWORKS 205
4 COMPARISON 207
5 CONCLUSION 209
Acknowledgments 209
References 209
Chapter 25. Relevant Explanations: Allowing Disjunctive Assignments 211
Abstract 211
1 INTRODUCTION 211
2 GIB EXPLANATION 212
3 GIB-MAP ALGORITHM 215
4 DISCUSSION 216
5 SUMMARY 216
References 218
Chapter 26. A Generalization of the Noisy-Or Model 219
Abstract 219
1 INTRODUCTION 219
2 THE GENERALIZED MODEL 220
3 CHARACTERIZING P(X/V) 221
4 INTERESTING SPECIAL CASES 221
5 COMPUTING P(X/U) 222
6 EXAMPLES 223
7 IMPLEMENTATION 226
Acknowledgements 226
References 226
Part 4: Automated Model Construction and Learning 228
Chapter 27. Using First-Order Probability Logic for the Construction of Bayesian Networks 230
Abstract 230
1 Introduction 230
2 Representing General ProbabilisticKnowledge 231
3 Representing Bayesian Networks 231
4 Simple Model Construction 232
5 More General Model Construction 233
6 Conclusions and Future Work 236
References 236
Chapter 28. Representing and Reasoning With Probabilistic Knowledge: A Bayesian Approach 238
Abstract 238
1 INTRODUCTION 238
2 REPRESENTING PROBABILISTIC KNOWLEDGE 238
3 PROBABILISTIC INFERENCE 242
4 RELATED WORK 243
5 CONCLUSIONS 245
References 245
Chapter 29. Graph-Grammar Assistance for Automated Generation of Influence Diagrams 246
Abstract 246
1 MODELING OF DECISIONS 246
2 GRAPH GRAMMARS 246
3 A GRAPH GRAMMAR FOR MEDICAL DECISIONS 248
4 DISCUSSION 249
Acknowledgments 251
References 251
Chapter 30.Using Causal Information and Local Measures to Learn Bayesian Networks 254
Abstract 254
1 Introduction 254
2 Learning Bayesian Networks 255
3 Localization of the Description Length 257
4 Incorporating Partial Domain Knowledge 258
5 Searching for the Best Constrained Network 258
6 Experiments 259
7 Refinement of Existent Networks 261
References 261
Chapter 31. Minimal Assumption Distribution Propagation in Belief Networks 262
Abstract 262
1 INTRODUCTION 262
2 PREVIOUS WORK 263
3 FRAMEWORK FOR LEARNING QBNS 263
4 PROPAGATING DISTRIBUTIONS 264
5 EXAMPLE 267
6 CONCLUSION 269
Acknowledgements 269
References 269
Chapter 32. An Algorithm for the Construction of Bayesian Network Structures from Data 270
Abstract 270
1 INTRODUCTION 270
2 MOTIVATION 271
3 DISCUSSION OF THE ALGORITHM 272
4 PRELIMINARY RESULTS 273
5 SUMMARY AND OPEN PROBLEMS 275
Acknowledgements 275
References 276
Chapter 33. A Construction of Bayesian Networks from Databases Based on an MDL Principle 277
Abstract 277
1 INTRODUCTION 277
2 DISCUSSION WITHOUTASSUMING BAYESIAN BELIEFNETWORKS 278
3 DISCUSSION ASSUMINGBAYESIAN BELIEF NETWORKS 280
4 Concluding Remarks 282
Acknowledgements 282
References 282
Appendix A: Proof of Theorem 2 283
Appendix .: Proof of Theorem 8 284
Chapter 34. Knowledge-Based Decision Model Construction for Hierarchical Diagnosis: A Preliminary Report 285
Abstract 285
1 Introduction 285
2 Skeleton of our Methodology 286
3 The Functional Component 286
4 The Bridge fault component 288
5 The Meta-Level Component 288
6 An Example 290
7 Related research 290
8 Conclusion 292
Acknowledgments 292
References 292
Part 5: Algorithms for Inference and Decision Making 294
Chapter 35. A Synthesis of Logical and Probabilistic Reasoning for Program Understanding and Debugging 296
Abstract 296
1 INTRODUCTION 296
2 PROBLEM FOCUS 296
3 OPERATING SYSTEMS AND PROGRAM UNDERSTANDING 297
4 LOGICAL ANALYSIS OF EXECUTION PATHS 298
5 UNCERTAINTY ABOUT ERRORS ON PATHS 299
6 UNCERTAINTY AND PATH IDENTIFICATION 301
7 SUMMARY 302
References 302
Chapter 36. An Implementation of a Method 
 
303 
Abstract 303
1 INTRODUCTION 303
2 STATISTICAL VARIANCE AND DIRICHLET DISTRIBUTIONS 304
3 THE APPROXIMATE PROPAGATIONMETHOD 304
4 THE MONTE CARLO INTEGRATIONMETHOD 305
5 COMPARISON OF METHODS 305
6 DISCUSSION AND CONCLUSION 306
References 307
Chapter 37. 
312 
Abstract 312
1 Introduction 312
2 Desiderata 312
3 Term Computation - a new task definition 313
4 Error Estimates 316
5 Making Term Computation Incremental 317
6 Experimental Evaluation 317
7 Discussion 318
8 Conclusion 318
References 318
Chapter 38. Deliberation Scheduling for Time-Critical Sequential Decision Making 320
Abstract 320
1 Introduction 320
2 Deliberation Scheduling 321
3 Precursor Deliberation 322
4 Recurrent Deliberation 324
5 Related Work and Conclusions 327
Acknowledgements 327
References 327
Chapter 39. Intercausal Reasoning with Uninstantiated Ancestor Nodes 328
Abstract 328
1 INTRODUCTION 328
2 QUALITATIVE PROBABILISTIC NETWORKS 329
3 UNINSTANTIATED ANCESTOR NODES 329
4 PRODUCT SYNERGY II 330
5 NOISY-OR DISTRIBUTIONS 332
6 CONCLUSION 333
APPENDIX: PROOFS 333
Acknowledgments 336
References 336
Chapter 40. Inference Algorithms for Similarity Networks 337
Abstract 337
1 INTRODUCTION 337
2 DEFINITION OF SIMILARITY NETWORKS 337
3 TWO TYPES OF SIMILARITY NETWORKS 339
4 INFERENCE USING SIMILARITY NETWORKS 340
5 INFERENTIAL AND DIAGNOSTIC COMPLETENESS 341
6 OPEN PROBLEMS 345
Acknowledgments 345
References 345
Chapter 41. Two Procedures for Compiling Influence Diagrams 346
ABSTRACT 346
1.0 INTRODUCTION 346
2.0 COMPILING DECISION NETWORKS 347
3.0 APPLICATIONS 351
4.0 FUTURE WORK 351
Acknowledgement 351
References 351
Chapter 42. An efficient approach for finding the MPE in belief networks 353
Abstract 353
1 Introduction 353
2 The algorithm for finding the MPE 354
3 Finding the / MPEs in belief networks 355
4 The MPE for a subset of variables in belief networks 358
5 Related work 359
6 Conclusion 360
References 360
Chapter 43. A Method for Planning Given Uncertain and Incomplete Information 361
Abstract 361
1 Introduction 361
2 State Representation 363
3 Reduction Operator 364
4 Representation of Plans 365
5 The Basic Planning Algorithm 366
6 Plan Reapplication 366
7 Super-Plans 367
8 Results 367
9 Conclusion 368
Acknowledgments 369
References 369
Chapter 44. heuse of conflicts in searching Bayesian networks 370
Abstract 370
1 Introduction 370
2 Background 371
3 Searching possible worlds 371
4 Estimating the Probabilities 372
5 Discussion 373
6 A Diagnosis Example 373
7 Search Strategy and Conflicts 375
8 Comparison with other systems 377
9 Conclusion 377
Acknowledgements 377
References 378
Chapter 45. GALGO: A Genetic ALGOrithm Decision Support Tool for ComplexUncertain Systems Modeled with Bayesian Belief Networks 379
Abstract 379
1 INTRODUCTION 379
2 BELIEF NETWORKS 379
3 GENETIC ALGORITHMS 380
4 THE ALGORITHM BEHIND GALGO 380
5 EXAMPLES 382
6 EXPERIMENTAL RESULTS 383
7 DISCUSSION 384
8. REFERENCES 386
Chapter 46. Using Tree-Decomposable Structures to ApproximateBelief Networks 387
Abstract 387
1 INTRODUCTION 387
2 BELIEF NETWORKS AND BELIEFTREES 387
3 OPTIMAL TREE-DECOMPOSABLENETWORKS 390
4 CONCLUSIONS AND FUTURERESEARCH 393
Acknowledgements 393
References 393
Chapter 47. Using Potential Influence Diagrams forProbabilistic Inference and Decision Making 394
Abstract 394
1 INTRODUCTION 394
2 CONDITIONAL INFLUENCEDIAGRAMS 394
3 POTENTIAL INFLUENCEDIAGRAMS 396
4 COMPOUND OPERATIONS ANDALGORITHMS 398
5 CONCLUSIONS AND FUTURERESEARCH 400
Acknowledgments 401
References 401
Chapter 48. Deciding Morality of Graphs is ...-complete 402
Abstract 402
1 INTRODUCTION 402
2 PRELIMINARIES 403
3 SUFFICIENT CONDITIONS FORMORALITY 403
4 NECESSARY CONDITIONS FORMORALITY 403
5 ANOTHER SUFFICIENTCONDITION 404
6 COMPLEXITY ANALYSIS 404
References 407
Chapter 49. Incremental computation of the value of perfect information instepwise-decomposable influence diagrams 411
Abstract 411
1 INTRODUCTION 411
2 STEPWISE-DECOMPOSABLEINFLUENCE DIAGRAMS 412
3 CONDENSING SDID'S 413
4 COMPUTING THE VALUE OFPERFECT INFORMATION 416
5 CONCLUSIONS 418
A cknowledgement 418
References 418
Part 6: Qualitative Reasoning 420
Chapter 50. Argumentative inference in uncertainand inconsistent knowledge bases 422
Abstract 422
1. Introduction 422
2. Arguments in Flat Knowledge bases 422
3. Properties of »— 424
4. Arguments in prioritized knowledgebases 426
5. Paraconsistent-Like Reasoning inLayered Knowledge Bases 428
6. Combining knowledge bases 429
7. Conclusion 430
References 430
Chapter 51. Argument Calculus and Networks 431
Abstract 431
1 INTRODUCTION 431
2 ARGUMENT DATABASES 431
3 INDEPENDENCE 433
4 ARGUMENT NETWORKS 434
5 COMPUTING ARGUMENTS 434
6 APPLICATIONS OFARGUMENT NETWORKS 435
CONCLUSION 438
ACKNOWLEDGEMENT 438
References 438
Chapter 52. Argumentation as a General Framework for Uncertain Reasoning 439
Abstract 439
1 INTRODUCTION 439
2 ARGUMENTATION 439
3 ARGUMENTATION CONSEQUENCERELATION 440
4 FLATTENING AND AGGREGATION 440
5 THE ARGUMENTATION THEOREMPROVER 440
6 AGGREGATION CRITERIA 442
7 QUANTITATIVE UNCERTAINTYCALCULI 442
8 ARGUMENTATION ANDDEFEASIBILITY 443
9 META-ARGUMENTATION 443
10 RESOLVING CONFLICTS ANDMAKING DECISIONS 443
11 CONCLUSION 444
Acknowledgements 444
References 444
Chapter 53. On reasoning in networks with qualitative uncertainty 446
Abstract 446
1 INTRODUCTION 446
2 A N EW QUALITATIVEAPPROACH 446
3 QUALITATIVE CHANGES INSIMPLE NETWORKS 448
4 A COMPARISON OF THETHREE FORMALISMS 449
5 QUALITATIVE CHANGES INMORE COMPLEX NETWORKS 449
6 INTEGRATION THROUGH QUALITATIVE CHANGE 451
7 DISCUSSION 452
8 SUMMARY 453
A cknowledgement s 453
References 453
Chapter 54. Qualitative Measures of Ambiguity 454
Abstract 454
1 Introduction 454
2 An Ambiguity Measure 455
3 Relationship between Ambiguityand Other Measures of Uncertainty 456
4 Conclusion 458
Appendix 458
References 461
Part 7: Interpretation and Comparisonof Approaches forReasoning Under Uncertainty 462
Chapter 55. A BAYESIAN VARIANT OF SHAFER'S COMMONALITIES FOR MODELLING UNFORESEEN EVENTS 464
Abstract 464
1 INTRODUCTION 465
2 MAPPING EVENTS INTO THE POWER SET OF F 466
3 EXPECTED UTILITY WITH NORMALIZEDCOMMONALITIES 467
4 CONCLUSIONS 469
Acknowledgements 470
References 470
Chapter 56. The Probability of a Possibility:Adding Uncertainty to Default Rules 472
Abstract 472
1 Introduction 472
2 Belief Revision and Possibilistic Logic 474
3 Counterfactual Probabilities 475
4 Generalized Imaging9 477
5 Concluding Remarks 478
Acknowledgements 479
References 479
Chapter 57. Possibilistic decreasing persistence 480
Abstract 480
1 Introduction 480
2 Background on possibilistic logic 481
3 Possibilistic decreasing persistence:the extrapolation problem 482
4 From qualitative to quantitativeaxioms for persistence schemata 484
5 Inferring nonmonotonic conclusionsfrom decreasing persistence 486
6 Concluding remarks 486
Acknowledgement 487
References 487
Chapter 58. DISCOUNTING AND COMBINATION OPERATIONS INEVIDENTIAL REASONING 488
Abstract 488
1 INTRODUCTION 488
2 HOW TO DISCOUNT THEEVIDENTIAL FUNCTIONS 488
3 DISCOUNTING A N DCOMBINATION OPERATIONS 490
4 SUMMARY 495
References 495
Chapter 59. Probabilistic Assumption-Based Reasoning 496
Abstract 496
1 MODELING PROPOSITIONALSYSTEMS 496
2 A LOGICAL V I EW OFSUPPORTS OF HYPOTHESES 497
3 COMPUTINGCHARACTERISTIC CLAUSES 499
4 COMPUTING DEGREES OFSUPPORT 500
5 CONCLUSION 502
References 502
Chapter 60. Partially Specified Belief Functions 503
Abstract 503
1 INTRODUCTION 503
2 BASIC CONCEPTS ANDRESULTS 504
3 THE MINIMUM SPECIFICITYAND LEAST COMMITMENTPRINCIPLES 506
4 THE FOCUSING PRINCIPLE 508
5 CONCLUSIONS 510
Acknowledgements 510
References 510
Chapter 61. Jeffrey's rule of conditioning generalized to belief functions. 511
Abstract: 511
1. Jeffrey's rule in probability theory. 511
2. Revision versus focusing 512
3. Jeffrey's rule applied to belieffunctions 514
4. Conclusions. 515
References. 516
Chapter 62. Inference with Possibilistic Evidence 517
Abstract 517
1 INTRODUCTION 517
2 POSSIBILISTIC EVIDENCE 518
3 INFERENCE WITHPOSSIBILISTIC EVIDENCE 520
4 SUMMARY AND DISCUSSION 524
References 525
Chapter 63. Constructing Lower Probabilities 526
Abstract 526
1 Introduction 526
2 An Alternative Construal of fi 526
3 Some Special Cases 527
4 Acknowledgements 529
References 529
Chapter 64. Belief Revision in Probability Theory 530
Abstract 530
1 INTRODUCTION 530
2 PROPAGATION VS. REVISION 530
3 EXPLICIT CONDITION VS. IMPLICITCONDITION 531
4 UPDATING VS. REVISION 533
5 A DEFECT OF THE BAYESIANAPPROACH 533
6 AN EXAMPLE 534
7 SUMMARY 536
Acknowledgements 537
References 537
Chapter 65. The Assumptions Behind Dempster's Rule 538
Abstract 538
1 INTRODUCTION 538
2 SOURCE STRUCTURES ANDBELIEF FUNCTIONS 538
3 COMBINATION RULES 539
4 DEMPSTER'S RULE OFCOMBINATION 540
5 BAYESIAN CONDITIONING 542
6 CONSTRAINTS ANDASSUMPTIONS ON C-RULES 542
7 DISCUSSION 543
Acknowledgements 544
References 544
Chapter 66. A Belief-Function Based Decision Support System 546
Abstract 546
1. INTRODUCTION 546
2. THEORETICAL BACKGROUND- TRANSFERABLE BELIEF MODEL 547
3. A BELIEF FUNCTION BASEDDECISION SUPPORT SYSTEM 547
4. AN EXAMPLE 548
3. CONCLUSIONS 552
Acknowledgments 552
References 552
Author Index 554

Produkt-Details

• Erscheint lt. Verlag: 12.5.2014
• Sprache: englisch
• Themenwelt: Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
• ISBN-10: 1-4832-1451-6 / 1483214516
• ISBN-13: 978-1-4832-1451-1 / 9781483214511