Human-Centered Artificial Intelligence

Research and Applications

1st Edition - May 15, 2022
Editors: Chang S. Nam, Jae-Yoon Jung, Sangwon Lee
Language: English
Paperback ISBN:
9 7 8 - 0 - 3 2 3 - 8 5 6 4 8 - 5
eBook ISBN:
9 7 8 - 0 - 3 2 3 - 8 5 6 4 9 - 2

Human-Centered Artificial Intelligence: Research and Applications presents current theories, fundamentals, techniques and diverse applications of human-centered AI. Sections addres… Read more

Purchase options

LIMITED OFFER

Save 50% on book bundles

Immediately download your ebook while waiting for your print delivery. No promo code is needed.

Institutional subscription on ScienceDirect

Request a sales quote

Human-Centered Artificial Intelligence: Research and Applications presents current theories, fundamentals, techniques and diverse applications of human-centered AI. Sections address the question, "are AI models explainable, interpretable and understandable?, introduce readers to the design and development process, including mind perception and human interfaces, explore various applications of human-centered AI, including human-robot interaction, healthcare and decision-making, and more. As human-centered AI aims to push the boundaries of previously limited AI solutions to bridge the gap between machine and human, this book is an ideal update on the latest advances.

Cover image
Title page
Table of Contents
Copyright
Contributors
Foreword
Preface
Part I. Frameworks of explainable AI
Chapter 1. Are AI models explainable, interpretable, and understandable?
1.1. Artificial intelligence: human and thinking machine
1.2. Explainability, interpretability, and understandability of AI
1.3. Why XAI is needed?
1.4. Categorization of XAI
Chapter 2. Explanation using model-agnostic methods
2.1. Introduction
2.2. Marginal effect of input feature
2.3. Contribution of each feature
2.4. Surrogate models
Appendix
Chapter 3. Explanation using examples
3.1. Introduction
3.2. Category of example-based explanations
3.3. Similarity-based methods
3.4. Influence-based methods
3.5. Case studies
3.6. Summary
Chapter 4. Explanation of ensemble models
4.1. Introduction
4.2. Ensemble models
4.3. Challenges of explaining ensemble models
4.4. Methods for interpreting ensemble models
4.5. Conclusions
Chapter 5. Explanation of deep learning models
5.1. Introduction
5.2. Activation-based models
5.3. Backpropagation-based models
Part II. User-centered AI design and development process
Chapter 6. AI as an explanation agent and user-centered explanation interfaces for trust in AI-based systems
6.1. Communication with computers: HCI and UX
6.2. Being with friends: new rationality and trust in companion AI
6.3. Explanation for trust: trustworthy AI and explainable AI
6.4. Explanation for results: AI as an explanation agent and explanation interfaces
Chapter 7. Anthropomorphism in human-centered AI: Determinants and consequences of applying human knowledge to AI agents
7.1. Introduction
7.2. Anthropomorphism: Using human knowledge for nonhuman targets
7.3. Anthropomorphism in human–AI interaction
7.4. Conclusion
Chapter 8. Designing a pragmatic explanation for the XAI system based on the user's context and background knowledge
8.1. Introduction
8.2. Explanation for the XAI system
8.3. Pragmatic explanation of van Fraassen
8.4. Summary and conclusion
Chapter 9. Interactive reinforcement learning and error-related potential classification for implicit feedback
9.1. Introduction
9.2. ErrP classification methods for implicit human feedback in RL
9.3. Interactive reinforcement learning
9.4. Discussion
9.5. Conclusion
Chapter 10. Reinforcement learning in EEG-based human-robot interaction
10.1. Introduction
10.2. The reinforcement learning problem
10.3. Reinforcement learning in EEG classification
10.4. Reinforcement learning using EEG in robot learning
10.5. Conclusions
Part III. Applications in human—AI interaction
Chapter 11. Shopping with AI: Consumers' perceived autonomy in the age of AI
11.1. Application of AI in advertising—its influence on consumers
11.2. Prospects and concerns for AI-based advertising
11.3. Challenges in AI-driven ads: a way to garner consumers' trust
11.4. Implications
Chapter 12. Use of deep learning techniques in EEG-based BCI applications
12.1. The electroencephalogram and brain–computer interfaces
12.2. Deep learning and EEGNet
12.3. Preparing the environment
12.4. Building and running the model
12.5. Understanding the model
12.6. Conclusions
Chapter 13. AI in human behavior analysis
13.1. Introduction
13.2. Human behavior analysis using AI
13.3. Sitting posture analysis using AI algorithms
13.4. Conclusion
Chapter 14. AI in nondestructive condition assessment of concrete structures: Detecting internal defects and improving prediction performance using prediction integration and data proliferation techniques
14.1. Introduction
14.2. Machine learning algorithms and applications
14.3. Discussion and conclusions
Part IV. Ethics, privacy, and policy in human—AI interaction
Chapter 15. Ethics of AI in organizations
15.1. Introduction
15.2. What are the principles of ethical AI?
15.3. Existing organizational theory
15.4. Integrating ethical principles of AI with organizational theory
15.5. Conclusion
Chapter 16. Designing XAI from policy perspectives
16.1. Introduction
16.2. Two psychological concerns in AI
16.3. Explainable AI
16.4. Remaining technical and political issues
16.5. Conclusion
Chapter 17. Responsible AI and algorithm governance: An institutional perspective
17.1. Introduction
17.2. Fair machine learning
17.3. Explainable machine learning
17.4. Conclusion
Author Index
Subject Index

Chang S. Nam

Chang S. Nam is currently a Professor of Industrial and Systems Engineering at North Carolina State University (NCSU), USA. He is also an associated faculty in the UNC/NCSU Joint Department of Biomedical Engineering, Department of Psychology, and Brain Research Imaging Center (BRIC) at UNC. He received a PhD at Virginia Tech. His research interests center around brain-computer interfaces, computational neuroscience, neuroergonomics, and human-AI/Robot/Automation interaction. He is the editor of “Brain-Computer Interfaces Handbook: Technological and Theoretical Advances” (with Drs. Nijholt and Lotte, CRC Press), “Neuroergonomics: Principles and Practices (Springer), “Mobile Brain-Body Imaging and the Neuroscience of Art, Innovation and Creativity (with Contreras-Vidal et al., Springer), “Trust in Human-Robot Interaction: Research and Applications” (with Lyons, Elsevier), and “Human-centered AI: Research and Applications” (with Jung & Lee, Elsevier). Currently, Nam serves as the Editor-in-Chief of the journal Brain-Computer Interfaces.

Affiliations and expertise

Professor, North Carolina State University, USA

Jae-Yoon Jung

Jae-Yoon Jung is a professor in the department of industrial and management systems engineering (IE) at Kyung Hee University (KHU), Korea, and also an adjunct professor of the department of the department of software convergence (SWCon), KHU. He is currently the director of Graduate Program, IE and Smart Factory Program at KHU. He is leading Industrial AI Lab at KHU. He received the Ph.D., M.S., and B.S. degrees in Industrial Engineering at Seoul National University (SNU), in 2005, 2001, and 1999, respectively. In SNU, he was supervised by prof. Suk-Ho Kang and Yeongho Kim in Intelligent Manufacturing Systems Lab. After that, he visited the Process Mining Group at Eindhoven University of Technology (TU/e) in the Netherland, supervised by prof. Wil van der Aalst. Before joining in KHU, he worked for u-Computing Innovation Center (uCIC), directed by Prof. Jinwoo Park, and he also studied in the Information Management Lab. at SNU, supervised by prof. Jonghun Park.

Affiliations and expertise

Professor, Department of Industrial and Management Systems Engineering (IE), Kyung Hee University (KHU), Korea; Adjunct professor, Department of the Department of Software Convergence (SWCon), KHU; Director, Graduate Program, IE and Smart Factory Program, KHU

Sangwon Lee

Sangwon Lee is an Associate Professor in Department of Interaction Science and Department of Applied Artificial Intelligence at Sugnkyunkwan University. He is also the director of ID square lab (Interaction Design and Development Laboratory). He received his BS degree from Korea University, and his MS degree and PhD degree from the Pennsylvania State University. His research interests lie in human-AI interaction, user experience, affective computing, user modelling, and explainable artificial intelligence.

Affiliations and expertise

Associate Professor, Department of Interaction Science and Department of Applied Artificial Intelligence, Sugnkyunkwan University, South Korea; Director, ID square lab (Interaction Design and Development Laboratory)