Table of Contents

Cover image

Title page

Copyright

About the Editors

Contributors

Preface

Introduction

1: Plan and Goal Recognition

1: Hierarchical Goal Recognition

1.1 Introduction

1.2 Previous Work

1.3 Data for Plan Recognition

1.4 Metrics for Plan Recognition

1.5 Hierarchical Goal Recognition

1.6 System Evaluation

1.7 Conclusion

2: Weighted Abduction for Discourse Processing Based on Integer Linear Programming

2.1 Introduction

2.2 Related Work

2.3 Weighted Abduction

2.4 ILP-based Weighted Abduction

2.5 Weighted Abduction for Plan Recognition

2.6 Weighted Abduction for Discourse Processing

2.7 Evaluation on Recognizing Textual Entailment

2.8 Conclusion

3: Plan Recognition Using Statistical–Relational Models

3.1 Introduction

3.2 Background

3.3 Adapting Bayesian Logic Programs

3.4 Adapting Markov Logic

3.5 Experimental Evaluation

3.6 Future Work

3.7 Conclusion

4: Keyhole Adversarial Plan Recognition for Recognition of Suspicious and Anomalous Behavior

4.1 Introduction

4.2 Background: Adversarial Plan Recognition

4.3 An Efficient Hybrid System for Adversarial Plan Recognition

4.4 Experiments to Detect Anomalous and Suspicious Behavior

4.5 Future Directions and Final Remarks

2: Activity Discovery and Recognition

5: Stream Sequence Mining for Human Activity Discovery

5.1 Introduction

5.2 Related Work

5.3 Proposed Model

5.4 Experiments

5.5 Conclusion

6: Learning Latent Activities from Social Signals with Hierarchical Dirichlet Processes

6.1 Introduction

6.2 Related Work

6.3 Bayesian Nonparametric Approach to Inferring Latent Activities

6.4 Experiments

6.5 Conclusion

3: Modeling Human Cognition

7: Modeling Human Plan Recognition Using Bayesian Theory of Mind

7.1 Introduction

7.2 Computational Framework

7.3 Comparing the Model to Human Judgments

7.4 Discussion

7.5 Conclusion

8: Decision-Theoretic Planning in Multiagent Settings with Application to Behavioral Modeling

8.1 Introduction

8.2 The Interactive POMDP Framework

8.3 Modeling Deep, Strategic Reasoning by Humans Using I-POMDPs

8.4 Discussion

8.5 Conclusion

4: Multiagent Systems

9: Multiagent Plan Recognition from Partially Observed Team Traces

9.1 Introduction

9.2 Preliminaries

9.3 Multiagent Plan Recognition with Plan Library

9.4 Multiagent Plan Recognition with Action Models

9.5 Experiment

9.6 Related Work

9.7 Conclusion

10: Role-Based Ad Hoc Teamwork

10.1 Introduction

10.2 Related Work

10.3 Problem Definition

10.4 Importance of Role Recognition

10.5 Models for Choosing a Role

10.6 Model Evaluation

10.7 Conclusion and Future Work

5: Applications

11: Probabilistic Plan Recognition for Proactive Assistant Agents

11.1 Introduction

11.2 Proactive Assistant Agent

11.3 Probabilistic Plan Recognition

11.4 Plan Recognition within a Proactive Assistant System

11.5 Applications

11.6 Conclusion

12: Recognizing Player Goals in Open-Ended Digital Games with Markov Logic Networks

12.1 Introduction

12.2 Related Work

12.3 Observation Corpus

12.4 Markov Logic Networks

12.5 Goal Recognition with Markov Logic Networks

12.6 Evaluation

12.7 Discussion

12.8 Conclusion and Future Work

13: Using Opponent Modeling to Adapt Team Play in American Football

13.1 Introduction

13.2 Related Work

13.3 Rush Football

13.4 Play Recognition Using Support Vector Machines

13.5 Team Coordination

13.6 Offline UCT for Learning Football Plays

13.7 Online UCT for Multiagent Action Selection

13.8 Conclusion

14: Intent Recognition for Human–Robot Interaction

14.1 Introduction

14.2 Previous Work in Intent Recognition

14.3 Intent Recognition in Human–Robot Interaction

14.4 HMM-Based Intent Recognition

14.5 Contextual Modeling and Intent Recognition

14.6 Experiments on Physical Robots

14.7 Discussion

14.8 Conclusion

Author Index

Subject Index