Path Analysis Models of an Autonomous Agent in a Complex Environment

  • Paul R. Cohen
  • David M. Hart
  • Robert St. Amant
  • Lisa A. Ballesteros
  • Adam Carlson
Conference paper
Part of the Lecture Notes in Statistics book series (LNS, volume 89)

Abstract

We seek explanatory models of how and why AI systems work in particular environments. We are not satisfied to demonstrate performance, we want to understand it. In terms of data and models, this means we are not satisfied with descriptive summaries, nor even with predictive models. We want causal models. In this brief abstract we will present descriptive, predictive and causal models of the behavior of agents that fight simulated forest fires. We will describe the shortcomings of descriptive and predictive models, and summarize path analysis, a common technique for inducing causal models.

Keywords

Covariance Dispatch 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [Asher 83]
    Asher, H.B. (1983) Causal Modeling. Sage Publications.Google Scholar
  2. [Cohen et al. 89]
    Cohen, PR., Greenberg, M.L., Hart, D.M., and Howe, A.E. (1989) “Trial by Fire: Understanding the Design Requirements for Agents in Complex Environments”, AI Magazine 10(3), 32–48.Google Scholar
  3. [Cohen et al. 93]
    Cohen, PR., Carlson, A., and Ballesteros, L.A. (1993) “Automating Path Analysis for Building Causal Models from Data”, Proc. 10th Intl. Conf. on Machine Learning, Amherst, MA, in press.Google Scholar
  4. [Cohen 94]
    Cohen, PR. (1994) Empirical Methods for Artificial Intelligence. The MIT Press, forthcoming.Google Scholar
  5. [Hart and Cohen 92]
    Hart, D.M. and Cohen, PR. (1992) “Predicting and Explaining Success and Task Duration in the Phoenix Planner”, Proc. 1st Intl. Conf. on AI Planning Systems, College Park, MD, 106–115.Google Scholar
  6. [Li 75]
    Li, C.C. (1975) Path Analysis - A Primer. Boxwood Press.Google Scholar
  7. [Loehlin 87]
    Loehlin, J.C. (1987) Latent Variable Models. Lawrence Erlbaum Associates.Google Scholar
  8. [Pearl and Verma 91]
    Pearl, J. and Verma, T.S. (1991) “A Theory of Inferred Causation”, Principles of Knowledge Representation and Reasoning: Proc. of 2nd Intl. Conf, Morgan Kaufmann, 441–452.Google Scholar
  9. [Spirtes 93]
    Spirtes, P., Glymour, C. and Scheines, R. (1993) Causation, Prediction and Search. Springer-Verlag, New York.MATHGoogle Scholar

Copyright information

© Springer-Verlag New York, Inc. 1994

Authors and Affiliations

  • Paul R. Cohen
    • 1
  • David M. Hart
    • 1
  • Robert St. Amant
    • 1
  • Lisa A. Ballesteros
    • 1
  • Adam Carlson
    • 1
  1. 1.Experimental Knowledge Systems Laboratory Department of Computer ScienceUniversity of MassachusettsAmherstUSA

Personalised recommendations