Advertisement

Managing multiple contexts efficiently

  • Gerhard K. Kraetzschmar
  • Josef Schneeberger
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1137)

Abstract

Assumption-based reasoning — using an ATMS — suffers from severe performance problems when the number of assumptions becomes very large. This is due to the exponential growth of ATMS labels. We suggest a new approach (FRMS) that avoids this problem by extending the RMS with context management functionality. This gives the problem solver explicit control over the contexts it actually needs.

Keywords

Reasoning Planning Agents and Multiagent Systems Type of Contribution: Poster 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [BKS94]
    C. Beckstein, G. K. Kraetzschmar, and J. Schneeberger. Distributed Plan Maintenance for Scheduling and Execution. In C. Bäckström and E. Sandewall, editors, Current Trends in AI Planning. IOS Press, 1994.Google Scholar
  2. [BLS92]
    M. Beetz, M. Lindner, and J. Schneeberger. Temporal Projection for Hierarchical, Partial-Order Planning. In Proc. of ISAI-92. AAAI Press, 1992.Google Scholar
  3. [dK86]
    J. de Kleer. An Assumption-based TMS. Artificial Intelligence, 28:127–224, 1986.Google Scholar
  4. [Gei94]
    T. Geisler. Ein anwendungsunabhängiges Unterstützungssystem zum integrierten annahmenbasierten und temporalen Schließen. Master's thesis, University of Erlangen, February 1994.Google Scholar
  5. [Kra96]
    G. K. Kraetzschmar. Distributed Reason Maintenance. PhD thesis, University of Erlangen, Erlangen, Germany, 1996.Google Scholar
  6. [KS96]
    G. K. Kraetzschmar and J. Schneeberger. FRMS: A focus-based reason maintenance system. Available via http from: http: //www. forwiss.uni-erlangen. de/∼ jws/papers/ki96-lv.ps, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • Gerhard K. Kraetzschmar
    • 1
  • Josef Schneeberger
    • 2
  1. 1.Neural Information Processing DepartmentUniversity of UlmUlmGermany
  2. 2.Knowledge Acquisition Research GroupFORWISSErlangenGermany

Personalised recommendations