Showing papers on "Crossover published in 1985"

Applying adaptive algorithms to epistatic domains

Abstract: John Holland has shown that when adaptive algorithms are used to search certain kinds of extremely large problem spaces, they will converge on a "good" solution fairly quickly. Such problem spaces are characterized by a low degree of epistasis. A host of classical search problems, however, are epistatic in nature. The present paper describes some new techniques for applying adaptive algorithms to epistatic domains, while retaining some of the strength of Holland's convergence proof. These techniques are described for two-dimensional bin-packing problems, and summarized for graph coloring problems. What makes these problems amenable to an adaptive approach is a two-stage evaluation procedure. Encodings of solutions are mutated and reproduced as they are in non-epistatic domains, but their evaluation is carried out after a decoding process. Using the techniques described, convergence is promoted in two ways: one of the natural mutation operators is a weaker version of Holland's crossover, and domain knowledge may be built into decoding processes so that the size of the search space is radically cut down.

Crossover functions by renormalization-group matching: Three-loop results

Abstract: Crossover functions are obtained to three-loop order in an expansion around dimension d =4 for the P theory. Expressions for the internal energy, susceptibility, and correlation length are given for the general n-component system in zero magnetic field and above the critical temperature. The full Helmholtz free energy, magnetization equation of state, and susceptibility are given for the Ising-like n ='1 case. Corrected values for the critical amplitude ratios as well as for the correctionto-scaling amplitude ratios are given to three-loop order. In this work, we extend earlier work' on crossover functions for thermodynamic functions in the critical region. These are obtained by a match point applied to the results of a conventional perturbation-theory expansion of the functions in question. Match-point methods were introduced by Nelson as a simplification of the differential generator approach to nonlinear. renormalization-group equations. As such, they appeared to be limited to firstorder, one-loop calculations. However, Bruce and Wallace showed that the same match-point idea could be used with field-theoretic perturbation methods and could thereby be carried out to any desired order. Applications of the approach also have been made by Theumann.

Scaling approach to phonon-fracton crossover.

Abstract: A scaling approach is formulated for the vibrational modes, from phonon to fracton, of a fractal network. The ratio of fracton-to-phonon density of vibrational states at crossover is found to be noncritical, i.e., independent of the crossover length scale. A steplike increase in the density of states at crossover is justified by an appeal to the normalization requirement on a fractal network. Applications are made to different fractal models. Recent effective-medium-approximation results are shown to violate scaling.

The Dimensionality Crossover in Quasi-1D Conductors

Abstract: A farmion path integral formalism is used for the study of one-dimensional (ID) gapless of Tomanaga-Luttinger (T-L) like electron gas models with interchain hopping. Scaling arguments and exact T-L1D results show that the one-particle crossover temperature TX is smaller than its free-electron gas value. Renormalized properties at TX and applications to (TMTSF)2X compounds are discussed.

Response change in linearized circuits and systems: Computational algorithms and applications

Abstract: A unified overview of linear and nonlinear system response change induced by multiparameter or state change is presented via an investigation of system equations transformed to a parameter based representation. General matrix modification formulas, applicable to analog and digital systems, are developed for both small and large change response sensitivities. Numerous previous response change results are examined and shown to be encompassed by the general approach taken here. Detailed analysis of numerical algorithms clearly demonstrates their relative cost efficiency for a large variety of applications. Exact efficiency crossover points are identified for both recursive and nonrecursive algorithms. Applications utilizing the parameter-based response change formulation are developed in several areas. They include pole assignment, feedback compensation, system augmentation and linkage, switched capacitor network analysis, nonlinear circuit analysis, and effective medium response of random variable systems.

Random-Field Critical Behavior and the Ginzburg Criterion

Abstract: The Ginzburg criterion for crossover from classical to nonclassical critical behavior is derived in the presence of random fields. Experiments on dilute antiferromagnets in a field are examined, and the possibility of such a crossover is investigated. We also point out that the (necessarily) symmetric logarithmic divergence observed in birefringence measurements can be accompanied by a discontinuity, and a fitting form which allows for this and other corrections to scaling is proposed.

Quantum-thermal crossover in the dissipative dynamics of the Sine-Gordon model

Abstract: The quantum-thermal crossover in the decay rate of metastable states of the quantum sine-Gordon model in dc field is discussed at low temperatures including dissipation in arbitrary dimension. In the low field limit, the thermal effect is always dominant, whereas in the higher field the quantum effect becomes important. The crossover field becomes larger by the dissipative effect. The crossover is sharp except the one dimensional underdamped case at low temperatures.

Crossover and noncrossover designs in four‐point parallel line analgesic assays

Abstract: Fifty-nine analgesic investigations designed as four-point parallel line crossover assays were examined. Sum of pain intensity differences (SPID) and total pain relief (TOTPAR) were the subjective response measures. Separate analyses with four-point crossover data and first-dose data (noncrossover) allowed comparison within each study of these two approaches. The crossover analysis allows for removal of the subject component of variance, which in these studies was a substantial fraction of the error variance (0.49 for SPID; 0.56 for TOTPAR). For this type of study, 2.4 times as many subjects would have to be recruited in a noncrossover design to obtain precision equivalent to that of the crossover design. Thus efficiency considerations argue for the crossover design in cases in which a treatment carryover effect may be assumed to be negligible. Clinical Pharmacology and Therapeutics (1985) 37, 242–252; doi:10.1038/clpt.1985.35

Program dependent crossover filter (PDC)

Abstract: An audio crossover filter which automatically tunes the crossover frequency according to the content of the program audio which is being filtered

Genetic analysis workshop III: A biological model

Abstract: A model for gene mapping that incorporates the concept of a limited crossover distribution is presented and applied to data simulated for GAW III.

Crossover behaviour of truncated Eden models

Abstract: Motivated by recent field-theory approaches to the Eden model of cluster growth, the authors study a truncated version with a maximal number of three nearest-neighbour particles. Numerical simulations on cubic lattices show that the modification is irrelevant to the asymptotic exponent and affects only the short-distance behaviour. Beyond a crossover scale, which increases with the dimensionality, the total mass of the clusters scales as the volume of a d-dimensional sphere with asymptotic filling factors of approximately 0.7 and approximately 0.5 in d=2 and d=3, respectively.

Kinetics of short-range ordering in metallic glasses : analysis of the crossover effect

Abstract: Model calculations were carried out for two different relaxation time spectra to simulate the relaxation curves that would be obtained for crossover experiments and other types of isothermal anneal sequences that give rise to inhomogeneous states of order. It is concluded that although the crossover effect clearly demonstrates the existence of more than a single relaxation process the response curve obtained in the crossover experiment tends to smear out the details of the relaxation time spectrum.