Interactive task learning : humans, robots, and agents acquiring new tasks through natural interactions / edited by Kevin A. Gluck, and John E. Laird ; Program Advisory Committee, Kenneth M. Ford [and six others].

Format:

Book

Contributor:

Gluck, Kevin A., editor.

Laird, John, 1954- editor.

Ford, Kenneth M., editor.

Series:

Strüngmann Forum reports

Language:

English

Subjects (All):

Human-computer interaction.

Artificial intelligence--Educational applications.

Artificial intelligence.

Intelligent tutoring systems.

Computer-assisted instruction.

Machine learning.

Robotics--Human factors.

Robotics.

Physical Description:

1 online resource (344 pages) : illustrations.

Other Title:

MIT Press CogNet.

Place of Publication:

Cambridge, Massachusetts : The MIT Press, [2018]

System Details:

text file

Summary:

Experts from a range of disciplines explore how humans and artificial agents can quickly learn completely new tasks through natural interactions with each other. Humans are not limited to a fixed set of innate or preprogrammed tasks. We learn quickly through language and other forms of natural interaction, and we improve our performance and teach others what we have learned. Understanding the mechanisms that underlie the acquisition of new tasks through natural interaction is an ongoing challenge. Advances in artificial intelligence, cognitive science, and robotics are leading us to future systems with human-like capabilities. A huge gap exists, however, between the highly specialized niche capabilities of current machine learning systems and the generality, flexibility, and in situ robustness of human instruction and learning. Drawing on expertise from multiple disciplines, this Strüngmann Forum Report explores how humans and artificial agents can quickly learn completely new tasks through natural interactions with each other. The contributors consider functional knowledge requirements, the ontology of interactive task learning, and the representation of task knowledge at multiple levels of abstraction. They explore natural forms of interactions among humans as well as the use of interaction to teach robots and software agents new tasks in complex, dynamic environments. They discuss research challenges and opportunities, including ethical considerations, and make proposals to further understanding of interactive task learning and create new capabilities in assistive robotics, healthcare, education, training, and gaming.

Contents:

1 Looking Forward to Interactive Task Learning p. 1 / Kevin A. Gluck and John E. Laird

2 Framing the Problem of Interactive Task Learning p. 9 / Tom M. Mitchell and Simon Garrod and John E. Laird and Stephen C. Levinson and Kenneth R. Koedinger

3 Functional Knowledge Requirements for Interactive Task Learning p. 19 / Robert E. Wray III and Niels A. Taatgen and Christian Lebiere and Katerina Pastra and Peter Pirolli and Paul S. Rosenbloom and Matthias Scheutz and Terrence C. Stewart and Janet Wiles

4 What People Learn from Instruction p. 53 / Christian Lebiere

5 An Ontological Perspective on Interactive Task Learning p. 63 / Charles Rich

6 The Representation of Task Knowledge at Multiple Levels of Abstraction p. 75 / Niels A. Taatgen

Interaction

7 Interaction for Task Instruction and Learning p. 91 / Andrea L. Thomaz and Elena Lieven and Maya Cakmak and Joyce Y. Chai and Simon Garrod and Wayne D. Gray and Stephen C. Levinson and Ana Paiva and Nele Russwinkel

8 Natural Forms of Purposeful Interaction among Humans: What Makes Interaction Effective? p. 111 / Stephen C. Levinson

9 Teaching Robots New Tasks through Natural Interaction p. 127 / Joyce Y. Chai and Maya Cakmak and Candace L. Sidner

10 The Essence of Interaction in Boundedly Complex, Dynamic Task Environments p. 147 / Wayne D. Gray and John K. Lindstedt and Catherine Sibert and Matthew-Donald D. Sangster and Roussel Rahman and Ropafadzo Denga and Marc Destefano

11 Task Instruction p. 169 / Julie A. Shah and Kevin A. Gluck and Tony Belpaeme and Kenneth R. Koedinger and Katharina J. Rohlfing and Han L. J. van der Maas and Paul Van Eecke and Kurt VanLehn and Anna-Lisa Vollmer and Matthew Yee-King

12 What Do Human Tutors Do? p. 193 / Kurt VanLehn

13 Strategies for Interactive Task Learning and Teaching p. 207 / Katrien Beuls and Luc Steels and Paul Van Eecke

14 Creativity and Feedback: Designing Systems to Support Student Learning and Improve Instruction p. 217 / Arthur Still and Matthew Yee-King and Mark d'Inverno

Learning New Tasks

15 Learning Task Knowledge p. 237 / Dario D. Salvucci and John E. Laird and Franklin Chang and Kenneth D. Forbus and Parisa Kordjamshidi and Tom M. Mitchell and Shiwali Mohan and Michael Spranger and Suzanne Stevenson and Andrea Stocco and J. Gregory Trafton

16 Early Developing Prerequisites for Human Interactive Task Learning p. 259 / Franklin Chang

17 Characteristics of the Learning Problem in Situated Interactive Task Learning p. 273 / John E. Laird and Shiwali Mohan and James Kirk and Aaron Mininger

Ethical Considerations

18 Ethical Aspects and Challenges for Interactive Task Learning p. 295 / Matthias Scheutz.

Notes:

OCLC-licensed vendor bibliographic record.

ISBN:

9780262349420

0262349426

OCLC:

1035389813

Access Restriction:

Restricted for use by site license.

Online:

OCLC metadata license agreement