×

Very fast simulated re-annealing. (English) Zbl 0681.90091

Summary: An algorithm is developed to statistically find the best global fit of a nonlinear nonconvex cost-function over a D-dimensional space. It is argued that this algorithm permits an annealing schedule for “temperature” T decreasing exponentially in annealing-time k, \(T=T_ 0\exp (-ck^{1/D})\). The introduction of re-annealing also permits to changing insensitivities in the multidimensional parameter-space. This annealing schedule is faster than fast Cauchy annealing, where \(T=T_ 0/k\), and much faster than Boltzmann annealing, where \(T=T_ 0/\ln k\). Applications are being made to fit empirical data to Lagrangians representing nonlinear Gaussian-Markovian systems.

MSC:

90C99 Mathematical programming
74A15 Thermodynamics in solid mechanics
PDF BibTeX XML Cite
Full Text: DOI Link

References:

[1] Jammer, M., The Philosophy of Quantum Mechanics (1974), Wiley: Wiley New York
[2] Metropolis, N.; Rosenbluth, A. W.; Rosenbluth, M. N.; Teller, A. H.; Teller, E., Equation of state calculations by fast computing machines, J. chem. Phys., 21, 1087-1092 (1953) · Zbl 1431.65006
[3] Kirkpatrick, S.; Gelatt, C. D.; Vecchi, M. P., Optimization by simulated annealing, Science, 220, 671-680 (1983) · Zbl 1225.90162
[4] German, S.; Geman, D., Stochastic relaxation, Gibbs distribution and the Baysian restorations in images, IEEE Trans. Patt. Anan. Mac. Int., 6, 721-741 (1984)
[5] Szu, H.; Hartley, R., Fast simulated annealing, Phys. Lett. A, 122, 157-162 (1987)
[6] Langouch, F.; Roekaerts, D.; Tirapegui, E., Functional Integration and Semiclassical Expansions (1982), Reidel: Reidel Dordrecht · Zbl 0505.28009
[7] Ingber, L., Statistical mechanics algorithm for response to targets (SMART), ((1985), AAAI-RCA: AAAI-RCA Melo Park, Calif), 258-264, Workshop on Uncertainty and Probability in Artificial Intelligence, UC Los Angeles, 14-16 August 1985
[8] Ingber, L., Statistical mechanics of nonlinear nonequilibrium financial markets, Mathl Modelling, 5, 343-361 (1984) · Zbl 0562.90018
[9] Ingber, L., \(C^3\) decision aids: statistical mechanics application of biological intelligence, (1987 Symposium on \(C^3\) Research. 1987 Symposium on \(C^3\) Research, National Defense University, Washington, D.C., 16-18 June 1987 (1987), National Defense University: National Defense University Washington, D.C), 49-57
[10] Ingber, L., Mathematical comparison of computer-models to exercise data: comparison of JANUS(T) to National Training Center data, (1988 JDL \(C^2\) Symposium. 1988 JDL \(C^2\) Symposium, Naval Postgraduate School, Monterey, Calif., 7-9 June 1988 (1988), SAIC: SAIC McLean, Va), 541-549
[11] Ingber, L., Riemannian contributions to short-ranged velocity-dependent nucleon-nucleoninteractions, Phys. Rev. D, 33, 3781-3784 (1986)
[12] Ingber, L., Statistical mechanics of neocortical interactions. Dynamics of synaptic modification, Phys. rev. A, 28, 395-416 (1983)
[13] Ingber, L., Statistical mechanics of neocortical interactions. Derivation of short-term-memory capacity, Phys. Rev. A, 29, 3346-3358 (1984)
[14] Ingber, L., Statistical mechanics of neocortical interactions. EEG dispersion relations, IEEE Trans. Biomed. Engng, 32, 91-94 (1985)
[16] Ingber, L., Noise-induced extrema in time-dependent Ginsburg-Landau systems, Mathl Modelling, 7, 525-528 (1986) · Zbl 0613.60075
[17] Ingber, L., Mesoscales in neocortex and in command, control and communications \((C^3)\) systems, (Delacour, J.; Levy, J. C.S., Systems with Learning and Memory Abilities: Proceedings, University of Paris, 15-19 June 1987 (1988), Elsevier: Elsevier Amsterdam), 387-409
[18] Nunez, P. L., Electric Fields of the Brain: The Neurophysics of EEG (1981), Oxford Univ. Press: Oxford Univ. Press New York
[19] Wehner, M. F.; Wolfer, W. G., Numerical evaluation of path integral solutions to Fokker-Planck equations—III. Time and functionally dependent coefficients, Phys. Rev. A, 35, 1795-1801 (1987)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.