Estimating information value in collaborative multi-agent planning systems

This paper addresses the problem of identifying the value of information held by a teammate on a distributed, multi-agent team. It focuses on a distributed scheduling task in which computer agents support people who are carrying out complex tasks in a dynamic environment. The paper presents a decision-theoretic algorithm for determining the value of information that is potentially relevant to schedule revisions, but is directly available only to the person and not the computer agent. The design of a "coordination autonomy" (CA) module within a coordination-manager system provided the empirical setting for this work. By design, the CA module depends on an external scheduler module to determine the specific effect of additional information on overall system performance. The paper describes two methods for reducing the number of queries the CA issues to the scheduler, enabling it to satisfy computational resource constraints placed on it. Experimental results indicate the algorithm improves system performance and establish the exceptional efficiency---measured in terms of the number of queries required for estimating the value of information---that can be achieved by the query-reducing methods.

[1]  Shlomo Zilberstein,et al.  A Value-Driven System for Autonomous Information Gathering , 2004, Journal of Intelligent Information Systems.

[2]  Eric Horvitz,et al.  Decision theory in expert systems and artificial intelligenc , 1988, Int. J. Approx. Reason..

[3]  David Sarne,et al.  Timing Interruptions for Better Human-Computer Coordinated Planning , 2006, AAAI Spring Symposium: Distributed Plan and Schedule Management.

[4]  Eric Horvitz,et al.  Models of attention in computing and communication , 2003, Commun. ACM.

[5]  Thomas Wagner,et al.  An Application View of COORDINATORS: Coordination Managers for First Responders , 2004, AAAI.

[6]  Laurence A. Kramer,et al.  Execution Monitoring and Replanning with Incremental and Collaborative Scheduling , 2005 .

[7]  Robert T. Clemen,et al.  Making Hard Decisions: An Introduction to Decision Analysis , 1997 .

[8]  Karen L. Myers,et al.  Active Coordination of Distributed Human Planners , 2002, AIPS.

[9]  Martha E. Pollack,et al.  Intelligent Technology for an Aging Population: The Use of AI to Assist Elders with Cognitive Impairment , 2005, AI Mag..

[10]  Jianhui Wu,et al.  Coordinated Plan Management Using Multiagent MDPs , 2006, AAAI Spring Symposium: Distributed Plan and Schedule Management.

[11]  Robin Cohen,et al.  A Decision Procedure for Autonomous Agents to Reason about Interaction with Humans , 2004 .

[12]  David Sarne,et al.  Sharing experiences to learn user characteristics in dynamic environments with sparse data , 2007, AAMAS '07.

[13]  Milind Tambe,et al.  A prototype infrastructure for distributed robot-agent-person teams , 2003, AAMAS '03.

[14]  Bart Selman,et al.  Constraint Programming for Distributed Planning and Scheduling , 2006, AAAI Spring Symposium: Distributed Plan and Schedule Management.

[15]  Robin Cohen,et al.  A User Modeling Approach to Determining System Initiative in Mixed-Initiative AI Systems , 2001, User Modeling.

[16]  Howard Raiffa,et al.  Applied Statistical Decision Theory. , 1961 .

[17]  Terry L. Zimmerman,et al.  Multi-Agent Management of Joint Schedules , 2006, AAAI Spring Symposium: Distributed Plan and Schedule Management.

[18]  Craig Boutilier,et al.  Who's asking for help?: a Bayesian approach to intelligent assistance , 2006, IUI '06.

[19]  John Phelps,et al.  GPGP – A Domain-Independent Implementation , 2006 .

[20]  D. J. White,et al.  Decision Theory , 2018, Behavioral Finance for Private Banking.

[21]  A. H. S. Ang,et al.  Probability Concepts in Engineering Planning and Design Volume II Decision , 1984 .

[22]  John Phelps,et al.  GPGP - The Doman-Independent Implementation , 2006, AAAI Spring Symposium: Distributed Plan and Schedule Management.

[23]  H. Raiffa,et al.  Applied Statistical Decision Theory. , 1961 .