Utilizing Natural Language for One-Shot Task Learning

Learning tasks from a single demonstration presents a significant challenge because the observed sequence is specific to the current situation and is inherently an incomplete representation of the procedure. Observation-based machine-learning techniques are not effective without multiple examples. However, when a demonstration is accompanied by natural language explanation, the language provides a rich source of information about the relationships between the steps in the procedure and the decision-making processes that led to them. In this article, we present a one-shot task learning system built on TRIPS, a dialogue-based collaborative problem solving system, and show how natural language understanding can be used for effective one-shot task learning.

[1]  Kathy Ryall,et al.  Learning hierarchical task models by defining and refining examples , 2001, K-CAP '01.

[2]  James F. Allen,et al.  TRIPS: An Integrated Intelligent Problem-Solving Assistant , 1998, AAAI/IAAI.

[3]  Stephanie Seneff,et al.  TINA: A Natural Language System for Spoken Language Applications , 1992, Comput. Linguistics.

[4]  Nathanael Chambers,et al.  Using Semantics to Identify Web Objects , 2006, AAAI.

[5]  Beth Ann Hockey,et al.  Grammar specialisation meets language modelling , 2002, INTERSPEECH.

[6]  Yolanda Gil,et al.  Acquiring Procedural Knowledge in EXPECT , 2000 .

[7]  W. Lewis Johnson,et al.  Learning domain knowledge for teaching procedural skills , 2002, AAMAS '02.

[8]  John R. Anderson,et al.  Learning to Act: Acquisition and Optimization of Procedural Skill , 2008 .

[9]  Acm Press,et al.  Proceedings of the ... International Joint Conference on Autonomous Agents and Multiagent Systems , 2002 .

[10]  Monica N. Nicolescu,et al.  Experience-based representation construction: learning from human and robot teachers , 2001, Proceedings 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems. Expanding the Societal Role of Robotics in the the Next Millennium (Cat. No.01CH37180).

[11]  Tessa A. Lau,et al.  Sheepdog: learning procedures for technical support , 2004, IUI '04.

[12]  Qiang Yang,et al.  Learning Actions Models from Plan Examples with Incomplete Knowledge , 2005, ICAPS.

[13]  Nate Blaylock,et al.  A problem solving model for collaborative agents , 2002, AAMAS '02.

[14]  Tessa A. Lau,et al.  Programming by demonstration: an inductive learning formulation , 1998, IUI '99.

[15]  Jim Blythe,et al.  Task learning by instruction in tailor , 2005, IUI.

[16]  John E. Laird,et al.  Learning procedural knowledge through observation , 2001, K-CAP '01.

[17]  James F. Allen,et al.  Linking Semantic and Knowledge Representations in a Multi-Domain Dialogue System , 2008, J. Log. Comput..

[18]  Dan Flickinger,et al.  An Open Source Grammar Development Environment and Broad-coverage English Grammar Using HPSG , 2000, LREC.