Uncertainty in Artificial Intelligence - Proceedings of the Seventh Conference on Uncertainty in Artificial Intelligence, UCLA, at Los Angeles, July 13-15, 1991

Front Cover

Uncertainty in Artificial Intelligence

Copyright Page

Table of Contents

Preface

Chapter 1. ARCO1: An Application of Belief Networks to the Oil Market

Abstract

1 Introduction

2 Domain Specifics

3 Model Variables

4 Scenarios

5 Forecasts

6 Conclusions

7 Acknowledgements

8 References

Chapter 2. "Conditional Inter-Causally Independent" node distributions, a property of "noisy-or" models

Abstract

1 EVIDENCE NODES THAT ARE COMMON TO MULTIPLE PARENTS

2 CONSTRUCTIVE SOLUTION OF THE BINARY VARIABLE INTER-CAUSAL DEPENDENCY

3 DISCUSSION

Acknowledgements

References

Chapter 3. Combining Multiple-valued Logics in Modular Expert Systems

Abstract

1 INTRODUCTION

2 ENTAILMENT SYSTEMS

3 A CLASS OF MULTIPLE-VALUED LOGICS FOR THE UNCERTAINTY MANAGEMENT IN RULE-BASED EXPERT SYSTEMS

4 INFERENCE PRESERVING MAPS BETWEEN MV-LOGICS

5 CONCLUSIONS AND FUTURE WORK

Acknowledgements

References

Chapter 4. Constraint Propagation with Imprecise Conditional Probabilities

Abstract

1 INTRODUCTION

2 STATEMENT OF THE PROBLEM

3 A LINEAR PROGRAMMING METHOD

4 GENERALIZED BAYES' THEOREM

5 LOCAL INFERENCE RULES

6 A CONSTRAINT PROPAGATION BASED ON INFERENCE RULES

7 AN EXAMPLE

8 CONJUNCTION AND DISJUNCTION

9 INDEPENDENCE ASSUMPTIONS

10 CONCLUSION

Acknowledgements

References

Chapter 5. BAYESIAN NETWORKS APPLIED TO THERAPY MONITORING

Abstract

1. INTRODUCTION

2. HIGH-LEVEL VIEW OF THE MODEL

3. INFERENCE

4. COMPUTING THE INFERENCES VIA STOCHASTIC SIMULATION

5. SPECIFIC MODEL FOR CYTOTOXIC CHEMOTHERAPY MONITORING IN BREAST CANCER

7. CONCLUSIONS

Acknowledgements

References

Chapter 6. Some Properties of Plausible Reasoning

Abstract

1 INTRODUCTION

2 NOTATION

3 THEORY

4 EXAMPLES

5 CONCLUSION

References

Chapter 7. Theory Refinement on Bayesian Networks

Abstract

1 Introduction

2 Bayesian Networks

3 Partial Bayesian networks

4 Representing alternative Bayesian networks

5 Theory Refinement

6 Extensions

7 Conclusion

Acknowledgements

References

Chapter 8. COMBINATION OF UPPER AND LOWER PROBABILITIES

Abstract

1 INTRODUCTION

2 'A PRIORI' INFORMATION

3 EVIDENTIAL INFORMATION

4 COMBINATION OF 'A PRIORI AND EVIDENTIAL INFORMATION

Acknowledgments

References

Chapter 9. A Probabilistic Analysis of Marker-Passing Techniques for Plan-Recognition

Abstract

1 Introduction

2 Probabilistic Schema Evaluation

3 Probabilistic Schema Selection

4 Path Calculations

5 Results

Acknowledgements

References

Chapter 10. Symbolic Probabilistic Inference with Continuous Variables

Abstract

1 Introduction

2 Overview of the SPI Algorithm

3 The SPI with Continuous Variables Algorithm

4 Conclusion

References

Chapter 11. Symbolic Probabilistic Inference with Evidence Potential

Abstract

1 Introduction

2 Evidence Potential Algorithm

3 Symbolic Inference with Evidence Potential

4 Examples

5 Conclusion

References

Chapter 12. A Bayesian Method for Constructing Bayesian Belief Networks from Databases

Abstract

1 INTRODUCTION

2 METHODS

3 PRELIMINARY RESULTS

4 SUMMARY OF THE LEARNING METHOD AND RELATED WORK

Acknowledgements

References

Chapter 13. Local Expression Languages for Probabilistic Dependence: a preliminary report

Abstract

1 Introduction

2 Overview of SPI

3 Local Expression Languages for Probabilistic Knowledge

4 Discussion

5 Conclusion

Acknowledgements

References

Chapter 14. Symbolic Decision Theory and Autonomous Systems

Abstract

1 INTRODUCTION

2 SYMBOLIC DECISION MAKING UNDER UNCERTAINTY

3 AUTONOMOUS DECISION MAKING UNDER UNCERTAINTY

Acknowledgements

References

Chapter 15. A REASON MAINTENANCE SYSTEM DEALING WITH VAGUE DATA

Abstract

INTRODUCTION

MANY-VALUED LOGICS AND RESOLUTION

DEFINITION OF A FUZZY TRUTH MAINTENANCE SYSTEM

CONCLUSION

Acknowledgements

References

Chapter 16. Advances in Probabilistic Reasoning

Abstract

1 Introduction

2 Representation and Inference

3 Knowledge Acquisition/Representation

4 Generalized Similarity Networks

5 Summary

References

Chapter 17. Probability Estimation in face of Irrelevant Information

Abstract

1 INTRODUCTION

2 THE UNDERLYING MODEL

3 THE ESTIMATION PROBLEM

4 JUSTIFICATION AND EXTENSIONS

5 COMPARISON TO OTHER WORK

6 CONCLUSION

Acknowledgments

References

Chapter 18. An Approximate Nonmyopic Computation for Value of Information

Abstract

1 INTRODUCTION

2 VALUE-OF-INFORMATION COMPUTATIONS FOR DIAGNOSIS

3 MYOPIC ANALYSIS

4 NONMYOPIC ANALYSIS

5 VALUE OF INFORMATION FOR A SUBSET OF EVIDENCE

6 RELAXATION OF THE ASSUMPTIONS

7 SUMMARY AND CONCLUSIONS

Acknowledgments

References

Chapter 19. Search-based Methods to Bound Diagnostic Probabilities in Very Large Belief Nets

Abstract

1 INTRODUCTION

2 QMR AND INTERNIST

3 QMR-BN: A PROBABILISTIC INTERPRETATION OF QMR

4 INFERENCE ALGORITHMS

5 NOTATION

6 RELATIVE PROBABILITY AND MARGINAL EXPLANATORY POWER

7 NEGATIVE PRODUCT SYNERGY AND THE MEP THEOREM

8 BOUNDS ON THE PROBABILITY OF EXTENSIONS

9 SEARCH METHOD

10 OBTAINING ABSOLUTE PROBABILITIES

11 PERFORMANCE OF TOPN

CONCLUSIONS

Acknowledgements

References

Chapter 20. Chapter Time-Dependent Utility and Action Under Uncertainty

Abstract

1 INTRODUCTION

2 A LIMITED REASONER

3 TIME-DEPENDENT UTILITY

4 PROTOS IN ACTION

5 SUMMARY

Acknowledgments

References

Chapter 21. Non-monotonic Reasoning and the Reversibility of Belief Change

Abstract

1 INTRODUCTION

2 BELIEF CHANGE AND INFERENCE

3 SEMANTICS FOR BELIEF CHANGE

4 ITERATED BELIEF CHANGE AND REVERSIBILITY

5 DISCUSSION

Acknowledgements

References

Chapter 22. Belief and Surprise - A Belief-Function Formulation

Abstract

1 INTRODUCTION

2 BELIEF FUNCTIONS AS A GENERAL FORMALIZATION MECHANISM

3 A CASE STUDY

4 DISCUSSION

5 CONCLUSION

Acknowledgements

Appendix - logical formulas and subsets of

References

Chapter 23. Evidential Reasoning in a Categorial Perspective: Conjunction and Disjunction of Belief Functions

Abstract

0 INTRODUCTION

1 FROM THE DYNAMICS OF BELIEFS TO CATEGORIES OR ... VICE VERSA

2 CATEGORIES OF "BELIEFS"

3 DISJUNCTIONS AND CONJUNCTIONS

4 COPRODUCTS AND CONJUNCTIONS

5 PRODUCTS AND DISJUNCTIONS

6 SEPARABLE BELIEF FUNCTIONS

7 CONCLUSIONS

Acknowledgments

References

Chapter 24. Reasoning with Mass Distributions

Abstract

1 INTRODUCTION

2 REPRESENTING KNOWLEDGE WITH MASS DISTRIBUTIONS

3 THE CONCEPT OF SPECIALIZATION

4 SPECIALIZATION MATRICES

5 CONCLUSIONS

Chapter 25. A Logic of Graded Possibility and Certainty Coping with Partial Inconsistency

ABSTRACT

1 INTRODUCTION

2 POSSIBILISTIC LOGIC : LANGUAGE AND SEMANTICS

3 AUTOMATED DEDUCTION IN POSSIBILISTIC LOGIC

CONCLUSION

Acknowledgements

References

Chapter 26. Conflict and Surprise: Heuristics for Model Revision

Abstract

1 INTRODUCTION

2 BACKGROUND

3 THEORETICAL FRAMEWORK

4 REBUTTALS

5 RARE CASES

6 DISCUSSION

Acknowledgements

References

Chapter 27. Reasoning under Uncertainty: Some Monte Carlo Results

Abstract

1 INTRODUCTION

2 METHOD

3 RESULTS

4 DISCUSSION

References

Chapter 28. Representation Requirements for Supporting Decision Model Formulation

Abstract

1 Introduction

2 An Example

3 The Decision Making Process

4 Summary of Inference Patterns and Representation Requirements

5 A Representation Design

6 Supporting General Inferences

7 Related Work

8 Discussion and Conclusion

Acknowledgments

References

Chapter 29. A Language for Planning with Statistics

Abstract

1 INTRODUCTION

2 KNOWLEDGE REPRESENTATION

3 INFERENCE

4 PLANNING

5 CONCLUSION

Acknowledgments

References

Chapter 30. A Modification to Evidential Probability

Abstract

1 Overview of the Problem

2 The Proposed Solution

3 Conclusions

Acknowledgments

References

Chapter 31. Investigation of Variances in Belief Networks

Abstract

1 INTRODUCTION

2 PRELIMINARY ASSUMPTIONS

3 DETERMINING THE VARIANCES IN INFERRED PROBABILITIES

4 OBTAINING AN UPPERBOUND FOR THE PRIOR VARIANCES

5 FUTURE RESEARCH

References

Chapter 32. A Sensitivity Analysis of Pathfinder: A Follow-up Study

Abstract

1 INTRODUCTION

2 DETAILS OF THE ANALYSIS

3 THE INITIAL STUDY

4 THE FOLLOW-UP STUDY

5 CONCLUSIONS

Acknowledgments

References

Chapter 33. Non-monotonic Negation in Probabilistic Deductive Databases

Abstract

1 Introduction

2 Syntax and Uses of General Probabilistic Logic Programs

3 Background: Fixpoint Theory for Pf-programs

4 Stability of Formula Functions

5 Stable Classes of Formula Functions

6 Discussion

7 Conclusions

Acknowledgements

References

Chapter 34. Management of Uncertainty in the Multi-Level Monitoring and Diagnosis of the Time of Flight Scintillation Array

Abstract

1 INTRODUCTION

2 BACKGROUND LITERATURE

3 TIME OF FLIGHT SCINTILLATION ARRAY

4 SYSTEM ARCHITECTURE

5 MANAGEMENT OFUNCERTAINTY AT THE MONITORING LEVEL

6 MANAGEMENT OF UNCERTAINTY AT THE STRUCTURAL REASONING LEVEL

7 MANAGEMENT OF UNCERTAINTY AT THE BEHAVIORAL REASONING LEVEL

8 IMPLEMENTATION

9 SUMMARY

Acknowledgements

References

Chapter 35. Integrating Probabilistic Rules into Neural Networks: A Stochastic EM Learning Algorithm

Abstract

1 INTRODUCTION

2 PROBABILISTIC NETWORKS

3 MAXIMUM LIKELIHOOD ESTIMATION

4 THE STOCHASTIC EM-ALGORITHM

5 DISCUSSION

Acknowledgements

References

Chapter 36. Representing Bayesian Networks within Probabilistic Horn Abduction

Abstract

1 Introduction

2 Probabilistic Horn Abduction

3 Representing Bayesian networks

4 Best-first abduction

5 Causation

6 Comparison with Other Systems

7 Conclusion

Acknowledgements

References

Chapter 37. DYNAMIC NETWORK UPDATING TECHNIQUES FOR DIAGNOSTIC REASONING

Abstract

1 INTRODUCTION

2 DYNAMICS OF DIAGNOSTIC REASONING UNDER UNCERTAINTY

3 SYSTEM ARCHITECTURE

4 MODEL CONSTRUCTION HEURISTICS

5 MODEL UPDATING

6 CONCLUSIONS

References

Chapter 38. High Level Path Planning with Uncertainty

Abstract

1 INTRODUCTION

2 U–GRAPH

3 PATH PLANNING WITH UNCERTAINTY

4 A FORMAL DEFINITION OF PATH PLANNING

5 RELATED WORK

6 CONCLUSION AND FUTURE WORK

Acknowledgements

References

Chapter 39. Formal Model of Uncertainty for Possibilistic Rules

OVERVIEW

1 POSSIBILITY DISTRIBUTIONS AND MEASURES

2 INFORMATION FUNCTIONS IN POSSIBILITY THEORY

3 DESIGN OF CONTINUOUS POSSIBILITY INFORMATION

4 PROPERTIES OF CONTINUOUS INFORMATION MEASURES

5 PRINCIPLE OF MAXIMUM UNCERTAINTY

REFERENCES

Chapter 40. Deliberation and its Role in the Formation of Intentions*

Abstract

1 INTRODUCTION

2 OVERVIEW

3 POSSIBLE WORLDS MODEL

4 DECISION TREES AND GOAL WORLDS

5 DELIBERATION AND INTENTIONS

6 CONCLUSIONS

References

Chapter 41. Handling Uncertainty during Plan Recognitionin Task-Oriented Consultation Systems

Abstract

1 INTRODUCTION

2 THE INFERENCE MECHANISM

3 THE PROBABILITY OF AN INTERPRETATION OF THE DISCOURSE

4 STRENGTH OF INFERENCES

5 INFORMATION CONTENT AND ITS USE

6 EXAMPLES

7 CONCLUSIONS

Acknowledgments

References

Chapter 42. TRUTH AS UTILITY: A CONCEPTUAL SYNTHESIS

Abstract

1 Introduction

2 Possible Worlds and Desirabilities

3 Desirability and Preference

4 Combination of Preference Functions

5 Possibility and Necessity

6 Preference, Similarity, and Fuzzy Logic

AckNowledgements

References

Chapter 43. PULCINELLAA General Tool for Propagating Uncertainty in Valuation Networks

Abstract

1. INTRODUCTION

2. THEORETICAL BACKGROUND

3. PULCINELLA

4. EXAMPLES

5. DISCUSSION

6. CONCLUSIONS

Acknowledgements

References

Chapter 44. Structuring Bodies of Evidence

Abstract

1 INTRODUCTION

2 BASIC NOTIONS IN EVIDENCE THEORY

3 PROPOSAL OF STRUCTURES

4 DEMPSTER RULE OF COMBINATION

5 LOCAL PROPAGATION OF INFORMATION

6 CONCLUSION

Acknowledgements

References

Chapter 45. On the Generation of Alternative Explanations with Implications for Belief Revision

Abstract

1 INTRODUCTION

2 CONSTRAINT SYSTEMS

3 GENERATING ALTERNATIVE EXPLANATIONS

4 BAYESIAN NETWORKS

5 DISCUSSION

Acknowledgments

References

Chapter 46. Completing Knowledge by Competing Hierarchies

Abstract

1 Introduction

2 The knowledge base

3· The control strategy

4. The application to a multi-hierarchical knowledge base

5. Discussion

Acknowledgements

References

Chapter 47. A Graph-Based Inference Method for Conditional Independence

Abstract

1. INTRODUCTION

2. NOTATION AND BASIC CONCEPTS

3. MULTIPLE UNDIRECTED GRAPHS

4. GRAPHICAL REPRESENTATION OF THE GRAPHOID AXIOMS

5. EXTENSIONS TO THE GRAPHICAL OPERATIONS

6. EXAMPLES

7. CONCLUSIONS

Acknowledgements

References

Chapter 48. A Fusion Algorithm for Solving Bayesian Decision Problems

Abstract

1 INTRODUCTION

2 A DIABETES DIAGNOSIS PROBLEM

3 VALUATION-BASED SYSTEM REPRESENTATION

4 SOLVING A VBS

5 A FUSION ALGORITHM

6 CONCLUSIONS

Acknowledgements

References

Chapter 49. Algorithms for Irrelevance-Based Partial MAPs

Abstract

1 INTRODUCTION

2 IB-MAP ALGORITHM

3 d-IB MAP ALGORITHM

4 FUTURE WORK

5 SUMMARY

Acknowledgements

References

Chapter 50. About Updating

Abstract

1. CONDITIONING RULES FOR BELIEF FUNCTIONS

2. THE SCENARIO: THE VOTING INTENTIONS STUDY

3. CONDITIONING

4. BELIEFS INDUCED BY THE PROPORTIONS

5. CONCLUSIONS

Acknowledgements

Bibliography

Chapter 51. Compressed Constraints in Probabilistic Logic and Their Revision

Abstract

1. PROLIFERATION OF WORLDS

2. OVERVIEW AND EXAMPLE

3. COMPRESSION USING KNOWLEDGE AND USING SEARCH

4. EXPRESSING THE CONSTRAINTS

5. REVISION WITH CONDITIONALS

6. AN EXAMPLE OF REVISION

7. REVISION USING POSTERIORS

8· CONCLUSIONS

Literature Cited

Chapter 52. Detecting Causal Relations in the Presence of Unmeasured Variables

Abstract

1 Introduction

2 Results

3 Can Theorem 3 Be Strengthened?

4 Appendix

Acknowledgements

References

Chapter 53. A Method for Integrating Utility Analysis into an Expert System for Design Evaluation under Uncertainty

Abstract

1. INTRODUCTION

2. INTEGRATION OF USER-DEFINED EVALUATION FUNCTION INTO EXPERT SYSTEM

3. EXAMPLE: AUTOMOTIVE BUMPER MATERIAL SELECTION KBS

4. CONCLUSIONS

Acknowledgment

References

Chapter 54. From Relational Databases to Belief Networks

Abstract

1 INTRODUCTION

2 RELATIONAL DATABASES

3 BELIEF NETWORKS

4 INITIAL DISTRIBUTIONS

5 CONCLUSIONS

Acknowledgement

References

Chapter 55. A Monte-Carlo Algorithm for Dempster-Shafer Belief

Abstract

1 INTRODUCTION

2 THE MONTE-CARLO ALGORITHM

3 COMPUTATION TIME

4 EXPERIMENTAL RESULTS

5 THE GENERALISED ALGORITHM

6 DISCUSSION

Acknowledgements

References

Chapter 56. Compatibility of Quantitative and Qualitative Representations of Belief

Abstract

1 INTRODUCTION

2 QUANTITATIVE BELIEF MEASURES

3 PREFERENCE RELATIONS VERSUS QUANTITATIVE BELIEF MEASURES

4 CONCLUSION

Acknowledgements

References

Chapter 57. An Efficient Implementation of Belief Function Propagation

Abstract

1 INTRODUCTION

2 SOME BASIC CONCEPTS ABOUT BELIEF FUNCTION NETWORKS

3 BELIEF FUNCTION PROPAGATION USING LOCAL COMPUTATION

4 A More Efficient Implementation

5 UPDATING MESSAGES

6 CONCLUSIONS

Acknowledgements

References

Chapter 58. A Non-Numeric Approach to Multi-Criteria/Multi-Expert Aggregation Based on Approximate Reasoning

Abstract

1. Introduction

2. PROBLEM FORMULATION

3. A Non-Numeric Technique Multi-Criteria Aggregation

4. Combining Expert's Opinions

5. CONCLUSION

6. REFERENCES

Chapter 59. Why Do We Need Foundations for Modelling Uncertainties?

1 What Are Foundations?

2 Do We Need Foundations At All?

3 Testability

4 Proliferation and Communication

5 Considering Foundations

6 What Are We Trying to Do?

7 What Are We Talking About?

8 Little but the Truth

9 More of the Truth

10 Usefulness

11 Practise and Theory

12 A Garden

Acknowledgments

References

Author Index