Abstract
Local minima make search and optimization harder. In this paper, we give a new global optimization approach, minimax aαβ relaxation, to cope with the pathological behavior of local minima. The minimax a αβ relaxation interplays a dual step minimax local to global optimization, an iterative local to global information propagation, and an adaptive local to global algorithm transition, within a parallel processing framework. In minimax αβ relaxation, a controls the rate of local to global information propagation and β controls the rate of algorithm transition from local to global optimization. Compared to the existing optimization approaches such as simulated annealing and local search, the minimax β relaxation demonstrates much better convergence performance for certain classes of constrained optimization problems.1
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
M.B. Clowes. On seeing things. Artificial Intelligence, 2: 79 - 116, 1971.
S.A. Cook. The complexity of theorem-proving procedures. In Proceedings of the Third ACM Symposium on Theory of Computing, pages 151 - 158, 1971.
M.R. Garey and D.S. Johnson. Computers and Intractability: A Guide to the Theory of NP-Completeness. W.H. Freeman and Company, San Francisco, 1979.
J. Gu. Optimization by multispace search. Technical Report UCECE-TR-90-001, Dept. of Electrical and Computer Engineering, Univ. of Calgary, Jan. 1990.
J. Gu. An a/?-relaxationfor global optimization. Technical Report UCECE-TR-91-003, Dept. of ECE, Univ. of Calgary, Apr. 1991.
J. Gu and X. Huang. A constraint network approach to a shape from shading analysis of a polyhedron. In Proceedings of IJCNN’92, pages 441 - 446, Beijing, Nov. 1992.
J. Gu. Multispace Search: A New Optimization Approach (Summary). In D.-Z. Du and X.-S. Zhang, editors, Lecture Notes in Computer Science, Vol. 834: Algorithms and Computation, pages 252 - 260. Springer-Verlag, Berlin, 1994.
J. Gu. Constraint-Based Search. Cambridge University Press, New York, 1995.
J. Gu, W. Wang, and T. C. Henderson. A parallel architecture for discrete relaxation algo¬rithm. IEEE Trans, on Pattern Analysis and Machine Intelligence, PAMI-9(6): 816 - 831, Nov. 1987.
J. Gu and W. Wang. A novel discrete relaxation architecture. IEEE Trans, on Pattern Analysis and Machine Intelligence, 14 (8): 857 - 865, Aug. 1992.
R. M. Haralick and L. G. Shapiro. The consistent labeling problem: Part 1. IEEE Trans, on Pattern Analysis and Machine Intelligence, PAMI-1 (2): 173 - 184, Apr. 1979.
B.K.P. Horn and B. G. Schunck. Determining optical flow. Technical Report Al Memo 572, Al Lab, MIT, Apr. 1978.
D.A. Huffman. Impossible Objects as Nonsense Sentences. In B. Meltzer and D. Michie, Eds., Machine Intelligence, pages 295–323. Edinburgh University Press, Edinburgh, Scotland, 1971.
A. K. Mackworth. Consistency in networks of relations. Artificial Intelligence, 8: 99 - 119, 1977.
A. Rosenfeld, R. A. Hummel, and S. W. Zucker. Scene labeling by relaxation operations. IEEE Trans, on Systems, Man, and Cybernetics, SMC-6(6): 420 - 433, June 1976.
R. V. Southwell. Relaxation Methods in Engineering Science. Oxford University Press, Lon¬don, 1940.
R. V. Southwell. Relaxation Methods in Theoretical Physics. Oxford University Press, London, 1946.
D. Waltz. Generating semantic descriptions from drawings of scenes with shadows. Technical Report AI271, MIT, Nov. 1972.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1995 Kluwer Academic Publishers
About this chapter
Cite this chapter
Gu, J. (1995). A Minimax αβ Relaxation for Global Optimization. In: Du, DZ., Pardalos, P.M. (eds) Minimax and Applications. Nonconvex Optimization and Its Applications, vol 4. Springer, Boston, MA. https://doi.org/10.1007/978-1-4613-3557-3_17
Download citation
DOI: https://doi.org/10.1007/978-1-4613-3557-3_17
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-3559-7
Online ISBN: 978-1-4613-3557-3
eBook Packages: Springer Book Archive