Adaptation Through Interpolation for Time-Critical
Case-Based Reasoning

N.Chatterjee and J.A.Campbell

Department of Computer Science
University College London
London WC1E 6BT
U.K.
email: {nchatter, jac}@uk.ac.ucl.cs



Abstract. The paper introduces and examines the relevance of the notion of
"interpolation" between case features, to facilitate fast adaptation of existing
cases to a current situation. When this situation is time-critical there is not enough
time for exhaustive comparison of various aspects of all the stored cases, so it
may not be possible to retrieve a high-quality match for a current problem within
a specified time-limit. Viewing imperfect adaptation as a process of interpolation
(or a set of possible processes with different qualities of interpolation) then gives
a robust and novel perspective for time-critical reasoning, as well as being
equally relevant for case-based reasoning (CBR) in general. Although
interpolation-like adaptation techniques have been used in some existing CBR
systems, they have not previously been treated explicitly from this perspective.
References

1.	Allen J.F. : Towards General Theory of Actions and Time. Artificial Intelligence, Vol 23,
No. 2, 1984, pp 123- 154.

2.	Barletta R. and Mark W. : Explanation-based Indexing of Cases. Proc. 7-th National
Conference on Artificial Intelligence. Saint Paul, Minnesota August 2 1-26, 1988, pp 541-
546.
3.	Campbell J.A. : Assessing the Quality of Knowledge-Based Systems: Lessons from the
EQUUS Project. Heuristics 3, 1991, pp 9-17.
4.	Chatterjee N. and Campbell J.A. : A Caching Scheme for Time-Critical and Case-Based
Reasoning. Proc. EXPERSYS-92, IITT-international, F-93460 Goumay sur Marne,
France, pp 477-482.
5.	Chatterjee N. and Campbell J.A. : A Caching Scheme for Time-Critical Knowledge-based
Computations. Proc. Sixth International Conference on Industrial and Engineering Applications 
of Artificial Intelligence and Expert Systems, Gordon and Breach Science Publishers SA., 1993, pp 61- 70.
6.	Conte S.D. and de Boor C. : Elementary Numerical Analysis : An Algorithmic Approach,
McGraw Hill, 1972.
7.	Converse T., Hammond K. and Marks M. : Leaming Modifications Rules from Expectation 
Failure. Proc. Case-Based Reasoning Workshop DARPA, Florida, May 31 - June 2.
1989, pp 110-114.

8.	Dodhiawala R., Sridharan N.S., Raulefs P. and Pickering C. : Real-Time AI Systems : A
Definition and An Architecture, Proc. IJCAI-89, Morgan Kaufmann, San Mateo, California, 1989, pp256-261.

9.	Hammond K.J. : Explaining and Repairing Plans that Fail. IJCAI-87, Milan, Italy, August
23-28,	1987, pp 109-114.

10.	Hammond K., Converse T. and Marks M . Learning from Opportunities. Storing and
Re-using Execution Time Optimizations Proc AAAI-88, Morgan Kaufmann, 1988, pp
536-540.

11.	Hinrichs T.R. and Kolodner J.L. : The Roles of Adaptation in Case-Based Design. Proc.
DARPA Case-Based Reasoning Workshop, Morgan Kaufmann 1991, pp 121-132.
12.	Kobayashi S. and Nakamura K. : Knowledge Compilation and Refinement for Fault Diagnosis. 
IEEE EXPERT, October 1991, pp 39-46.

13.	Kolodner J.L : "An Introduction to Case-Based Reasoning", Artificial Intelligence Review,
vol. 6, No. 1, 1992, pp 3-34.

14.	Sycara K. : Using Case-Based Reasoning for Plan Adaptation and Repair. Proc. Case-Based 
Reasoning Workshop DARPA, May 10-13, 1988 , Clearwater beach, Florida, pp
425-434.

15.	Wolstencroft J. : Restructuring, Reminding and Repair: Whats Missing from Models of
Analogy?, AI Communications, vol. 2, 1989, pp 58-7 1.

16.	Zadeh L.A. : A Theory in Approximate Reasoning in Machine Intelligence 9, eds J.E.
Hayes, D. Michie and L.I. Mikulich. New York, Halstead Press pp 149-194, 1979.
