This Article 
 Bibliographic References 
 Add to: 
SHAPES: A Novel Approach for Learning Search Heuristics in Under-Constrained Optimization Problems
September-October 1997 (vol. 9 no. 5)
pp. 731-746

Abstract—Although much research in machine learning has been carried out on acquiring knowledge for problem-solving in many problem domains, little effort has been focused on learning search-control knowledge for solving optimization problems. This paper reports on the development of SHAPES, a system that learns heuristic search guidance for solving optimization problems in intractable, under-constrained domains based on the Explanation-Based Learning (EBL) framework. The system embodies two new and novel approaches to machine learning. First, it makes use of explanations of varying levels of approximation as a mean for verifying heuristic-based decisions, allowing heuristic estimates to be revised and corrected during problem-solving. The provision of such a revision mechanism is particularly important when working in intractable and under-constrained domains, where heuristics tend to be highly over-generalized, and hence at times will give rise to incorrect results. Second, it employs a new linear and quadratic programming-based weight-assignment algorithm formulated to direct search toward optimal solutions under best-first search. The algorithm offers a direct method for assigning rule strengths and, in so doing, avoids the need to address the credit-assignment problem faced by other iterative weight-adjustment methods.

[1] N. Bhatnagar and J. Mostow, "Adaptive Search by Explanation-Based Learning of Heuristic Censors," Proc. Eighth Nat'l Conf. Artificial Intelligence, pp. 895-901, Morgan Kaufmann, 1990.
[2] W.W. Cohen, "Abductive Explanation-Based Learning: A Solution to the Multiple Inconsistent Explanation Problem," Machine Learning, vol. 8, pp. 167-219, 1992.
[3] G.B. Dantzig Linear Programming and Extensions. Princeton Univ. Press, 1963.
[4] G. DeJong and R. Mooney, "Explanation-Based Learning: An Alternative View," Machine Learning, vol. 1, pp. 145-176, 1986.
[5] K.P. Doan and K.P. Wong, "SHAPES: A System for Learning Search Heuristics for Under-Constrained Optimization Problems," Technical Report No. AIPS-12-93-01, Dept. of Electrical and Electronic Engineering, Univ. of Western Australia, 1993.
[6] R.J. Doyle, "Constructing and Refining Causal Explanations from an Inconsistent Domain Theory," Proc. Fifth Nat'l Conf. Artificial Intelligence, pp. 538-544, Morgan Kaufmann, 1986.
[7] T. Ellman, "Approximate Theory Formation: An Explanation-Based Approach," Proc. Seventh Nat'l Conf. Artificial Intelligence, pp. 570-574, Morgan Kaufmann, 1988.
[8] T. Ellman, "Explanation-Based Learning: A Survey of Programs and Perspectives," ACM Computing Surveys, vol. 21, pp. 163-220, 1989.
[9] T. Fawcett and P. Utgoff, "A Hybrid Method for Feature Generation," Proc. Ninth Int'l Conf. Machine Learning, pp. 137-141, Morgan Kaufmann, 1992.
[10] H. Habibollahzadeh and J. Bubenko, "Application of Decomposition Techniques to Short-Term Operation Planning of Hydrothermal Power System," IEEE Trans., vol. PWRS-1, pp. 41-47, 1986.
[11] J.H. Holland, Adaptation in Natural and Artificial Systems. Univ. of Michigan Press, 1975.
[12] J.H. Holland, "Escaping Brittleness: The Possibilities of General-Purpose Learning Algorithms Applied to Parallel Rule-Based Systems," Machine Learning: An Artificial Intelligence Approach, vol.2, R. Michalski, J. Carbonell, and T. Mitchell, eds, pp. 593-623, Morgan Kaufmann, 1986.
[13] S. Kedar-Cabelli, "Toward a Computational Model of Purpose-Directed Analogy," Readings in Machine Learning, J. Shavlik and T. Dietterich, eds., pp. 647-656, Morgan Kaufmann, 1990.
[14] R.M. Keller, "Concept Learning in Context," Proc. Fourth Int'l Workshop Machine Learning, pp. 91-102, Morgan Kaufmann, 1987.
[15] S. Kirkpatrick, C.D. Gelatt, and M.P. Vecchi, "Optimisation by Simulated Annealing," Science, vol. 220, pp. 671-680, 1983.
[16] J. Laird, P. Rosenbloom, and A. Newell, "Chunking in SOAR: The Anatomy of a General Learning Mechanism," Machine Learning, vol. 1, pp. 11-46, 1986.
[17] K. Lee and S. Mahajan, “A Pattern Classification Approach to Evaluation Function Learning,” Artificial Intelligence, vol. 36, no. 1, pp. 1-25, 1988.
[18] S. Minton, Learning Search Control Knowledge: An Explanation-Based Approach. Kluwer Academic, 1988.
[19] T. Mitchel, R. Keller, and S. Kedar-Cabelli, “Explanation-Based Generalization: A Unifying View,” Machine Learning, pp. 47–80, 1986.
[20] T. Mitchell, S. Mabadevan, and L. Steinberg, "LEAP: A Learning Apprentice for VLSI Design," Machine Learning: An Artificial Intelligence Approach, vol. 3, Y. Kodratoff and R. Michalski, eds., pp. 271-289, Morgan Kaufmann, 1990.
[21] S. Mokhtari, J. Singh, and B. Wollenberg, "A Unit Commitment Expert System," IEEE Trans. Power Systems, Vol. 3, No. 1, Feb. 1988, pp. 272-277.
[22] J. Mostow, "Machine Transformation of Advice into a Heuristic Search Procedure," Machine Learning: An Artificial Intelligence Approach," R. Michalski, J. Carbonell, and T. Mitchell, eds., pp. 367-403, Morgan Kaufmann, 1983.
[23] J. Mostow and A. Prieditis, "Discovering Admissible Heuristics by Abstracting and Optimizing," Proc. 11th Int'l Joint Conf. Artificial Intelligence, Morgan Kaufmann, 1989.
[24] J. Muckstadt and S. Koenig, "An Application of Lagrangian Relaxation to Scheduling in Power-Generation Systems," Operation Research, vol. 23, pp. 387-403, 1977.
[25] J. Muckstadt and R. Wilson, "An Application of Mixed-Integer Programming Duality to Scheduling Thermal Generating Systems," IEEE Trans., vol. PAS-87, pp. 1,968-1,978, 1968.
[26] A.E. Prieditis, "Machine Discovery of Effective Admissible Heuristics," Machine Learning, vol. 12, pp. 117-141, 1993.
[27] D.EE. Rumelhart and J.L. McClelland, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, MIT Press, Cambridge, Mass., 1986.
[28] A.L. Samuel, "Some Studies in Machine Learning Using the Game of Checkers," IBM J., vol. 3, pp. 210-229, 1959.
[29] A.L. Samuel, "Some Studies in Machine Learning Using the Game of Checkers II," IBM J., vol. 11, pp. 601-617, 1967.
[30] W. Snyder Jr., H. Powell, and J. Rayburn, "Dynamic Programming Approach to Unit Commitment," IEEE Trans., vol. PWRS-2, pp. 339-350, 1987.
[31] A. Wood and B. Wollenberg, Power Generation, Operation, and Control, John Wiley&Sons, 1984.
[32] P. Utgoff and S. Saxena, "Learning a Preference Predicate," Proc. Fourth Int'l Workshop Machine Learning, pp. 115-121, Morgan Kaufmann, 1987.
[33] P. Winston, T. Binford, B. Katz, and M. Lowry, "Learning Physical Descriptions from Functional Definitions, Examples, and Precedents," Proc. Nat'l Conf. Artificial Intelligence, pp. 433-439, Morgan Kaufmann, 1983.
[34] A.K. Jain and R.C. Dubes, Algorithms for Clustering Data. Englewood Cliffs, N.J.: Prentice Hall, 1988.
[35] K.P. Wong and K.P. Doan, "Artificial Intelligence Algorithm for Daily Scheduling of Thermal Generators," IEE Proc. Part C, vol. 138, pp. 518-534, 1991.
[36] K.P. Wong and Y.W. Wong, "Thermal Generator Scheduling Using Hybrid Genetic/Simulated-Annealing Approach," IEE Proc. Gener. Transmission Distribution, vol. 142, no. 4, 1995.
[37] F. Zhuang and F.D. Galiana, "Unit Commitment by Simulated Annealing," IEEE Trans., vol. PWRS-5, pp. 311-317, 1990.

Index Terms:
Explanation-Based Learning, heuristics, intractability, search, weight assignment.
Khanh P.V. Doan, Kit Po Wong, "SHAPES: A Novel Approach for Learning Search Heuristics in Under-Constrained Optimization Problems," IEEE Transactions on Knowledge and Data Engineering, vol. 9, no. 5, pp. 731-746, Sept.-Oct. 1997, doi:10.1109/69.634752
Usage of this product signifies your acceptance of the Terms of Use.