The use of heuristics such as Genetic Algorithm optimisation methods is appealing in a large range of inverse problems. The problem presented here deals with the mathematical analysis of sequences generated by finite automata. There is no known general exact method for solving the associated inverse problem. GA optimisation techniques can provide useful results, even in the very particular area of mathematical analysis. This paper presents the results we have obtained on the inverse problem for fixed point automata. Software implementation has been developed with the help of “ALGON”, our home-made Genetic Algorithm software.

We present a new approach in order to improve the convergence of Monte Carlo (MC) simulations of molecular systems belonging to complex energetic landscapes: the problem is redefined in terms of the dynamic allocation of MC move frequencies depending on their past efficiency, measured with respect to a relevant sampling criterion. We introduce various empirical criteria with the aim of accounting for the proper convergence in phase space sampling. The dynamic allocation is performed over parallel simulations by means of a new evolutionary algorithm involving 'immortal' individuals. The method is bench marked with respect to conventional procedures on a model for melt linear polyethylene. We record significant improvement in sampling efficiencies, thus in computational load, while the optimal sets of move frequencies are liable to allow interesting physical insights into the particular systems simulated. This last aspect should provide a new tool for designing more efficient new MC moves.

We present the first application of Genetic Algorithms to the analysis of data from an aperiodically ordered system, high resolution X-Ray diffraction spectra from multilayer heterostructures arranged according to a deterministic or random scheme. This method paves the way to the solution of the “inverse problem”, that is the retrieval of the generating disorder from the investigation of the spectra of an unknown sample having non crystallographic, non quasi-crystallographic order.

When considering noisy fitness functions for some CPU-time consuming applications, a trade-off problem arise: how to reduce the influence of the noise while not increasing too much computation time. In this paper, we propose and experiment some new strategies based on an exploitation of historical information on the algorithm evolution, and a non-generational evolutionary algorithm.

We present a new approach to performing molecular simulations using evolutionary algorithms. The main application is the simulation of dense amorphous polymers and the goal is to improve the efficiency of sampling, in other words to obtain valid samples from the phase state more rapidly. Our approach is based on parallel Markovian Monte Carlo simulations of the same physico-chemical system, where we optimise some Monte Carlo parameters by means of a real coded genetic algorithm

Presents a theoretical analysis that relates an irregularity measure of a fitness function to the so-called GA (genetic algorithm) deception. This approach is a continuation of the work of Lutton and Le/spl acute/vy-Ve/spl acute/hel (1995) that has presented a deception analysis of Ho/spl uml/lder functions. The analysis developed in this paper is a generalization of this work in two ways: first, we use a "bitwise regularity" instead of a Ho/spl uml/lder exponent as a basis for our deception analysis; and second, we perform a similar deception analysis of a GA with uniform crossover. We finally propose to use the bitwise regularity coefficients in order to analyse the influence of chromosome encoding on the GA efficiency, and we present experiments with Gray encoding.