Mixed initiative interfaces for learning tasks: SMARTedit talks back

ABSTRACT

Applications of machine learning can be viewed as teacher-student interactions in which the teacher provides training examples and the student learns a generalization of the training examples. One such application of great interest to the IUI community is adaptive user interfaces. In the traditional learning interface, the scope of teacher-student interactions consists solely of the teacher/user providing some number of training examples to the student/learner and testing the learned model on new examples. Active learning approaches go one step beyond the traditional interaction model and allow the student to propose new training examples that are then solved by the teacher. In this paper, we propose that interfaces for machine learning should even more closely resemble human teacher-student relationships. A teacher's time and attention are precious resources. An intelligent student must proactively contribute to the learning process, by reasoning about the quality of its knowledge, collaborating with the teacher, and suggesting new examples for her to solve. The paper describes a variety of rich interaction modes that enhance the learning process and presents a decision-theoretic framework, called DIAManD, for choosing the best interaction. We apply the framework to the SMARTedit programming by demonstration system and describe experimental validation and preliminary user feedback.

REFERENCES

• 1: Dana Angluin, Queries and Concept Learning, Machine Learning, v.2 n.4, p.319-342, April 1988 [doi>10.1023/A:1022821128753]
• 2: Peter Auer, An Improved On-line Algorithm for Learning Linear Evaluation Functions, Proceedings of the Thirteenth Annual Conference on Computational Learning Theory, p.118-125, June 28-July 01, 2000
• 3: Mathias Bauer , Dietmar Dengler , Gabriele Paul, Instructible information agents for Web mining, Proceedings of the 5th international conference on Intelligent user interfaces, p.21-28, January 09-12, 2000, New Orleans, Louisiana, United States [doi>10.1145/325737.325758]
• 4: Mihai Boicu , Gheorghe Tecuci , Dorin Marcu , Michael Bowman , Ping Shyr , Florin Ciucu , Cristian Levcovici, Disciple-COA: From Agent Programming to Agent Teaching, Proceedings of the Seventeenth International Conference on Machine Learning, p.73-80, June 29-July 02, 2000
• 5: David Cohn , Les Atlas , Richard Ladner, Improving Generalization with Active Learning, Machine Learning, v.15 n.2, p.201-221, May 1994 [doi>10.1023/A:1022673506211]
• 6: Allen Cypher, Eager: programming repetitive tasks by demonstration, Watch what I do: programming by demonstration, MIT Press, Cambridge, MA, 1993
• 7: Dan Diaper, Task Analysis for Human-Computer Interaction, Prentice Hall PTR, Upper Saddle River, NJ, 1990
• 8: G. Ferguson and J. Allen. Arguing about plans: Plan representation and reasoning in mixed-initiative planning. In Proceedings of the Second Int'l. Conf. on Arti~cial Intelligence Planning Systems, pages 43{48. Menlo Park, Calif.: AAAI Press, 1994.
• 9: G. Ferguson, J. Allen, and B. Miller. TRAINS-95: Towards a mixed-initiative planning assistant. In Proceedings of the Third Int'l. Conf. on Arti~cial Intelligence Planning Systems, pages 70{77, Edinburgh, Scotland, May 1996. Menlo Park, Calif.: AAAI Press.
• 10: Tessa A. Lau , Pedro Domingos , Daniel S. Weld, Version Space Algebra and its Application to Programming by Demonstration, Proceedings of the Seventeenth International Conference on Machine Learning, p.527-534, June 29-July 02, 2000
• 11: Tessa A. Lau , Daniel S. Weld, Programming by demonstration: an inductive learning formulation, Proceedings of the 4th international conference on Intelligent user interfaces, p.145-152, January 05-08, 1999, Los Angeles, California, United States [doi>10.1145/291080.291104]
• 12: Neal Lesh , Charles Rich , Candace L. Sidner, Using plan recognition in human-computer collaboration, Proceedings of the seventh international conference on User modeling, p.23-32, June 1999, Banff, Canada
• 13: Pattie Maes and Robyn Kozierok. Learning interface agents. In Proceedings of the Fourteenth National Conf. on Arti~cial Intelligence, pages 459{465, 1993.
• 14: David Maulsby , Ian H. Witten, Metamouse: an instructible agent for programming by demonstration, Watch what I do: programming by demonstration, MIT Press, Cambridge, MA, 1993
• 15: Thomas M. Mitchell, Machine Learning, McGraw-Hill Higher Education, 1997
• 16: Brad A. Myers, Peridot: creating user interfaces by demonstration, Watch what I do: programming by demonstration, MIT Press, Cambridge, MA, 1993
• 17: Howard Rai~a. Decision Analysis: Introductory Lectures on Choices Under Uncertainty. Addison-Wesley, 1968.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE

Additional Classification:
  H. Information Systems
  H.1 MODELS AND PRINCIPLES
  H.4 INFORMATION SYSTEMS APPLICATIONS
      H.4.2 Types of Systems
          Subjects: Decision support (e.g., MIS)

  K. Computing Milieux
  K.3 COMPUTERS AND EDUCATION

General Terms:
Design, Human Factors, Management, Measurement, Performance, Theory

Keywords:
machine learning applications, mixed initiative, programming by demonstration