Tytuł artykułu
Autorzy
Identyfikatory
Warianty tytułu
Dynamics of adaptation of phenotypic evolution methods
Języki publikacji
Abstrakty
W dysertacji przedstawiono rezultaty analizy teoretycznej metody poszukiwań ewolucyjnych z miękką selekcją. Metoda bazuje na prostym modelu darwinowskiej ewolucji fenotypowej działającej z selekcją proporcjonalną i mutacją gaussowską (z rozkładem normalnym). Połączenie obu mechanizmów reprodukcji powoduje, że ewoluujące populacje są w stanie eksplorować nieograniczone wielowymiarowe rzeczywiste przestrzenie w poszukiwaniu wyższych przystosowań funkcji celu. Algorytmy optymalizacyjne implementujące metodę mają pożądane w optymalizacji właściwości: zdolność przekraczania siodeł pomiędzy optimami funkcji celu, a tym samym znajdowania optimów globalnych, niewielki spadek efektywności wraz ze wzrostem wymiarowości przestrzeni poszukiwań oraz małą wrażliwość na niedokładności i zaszumienia w odczycie wartości funkcji celu. Zaprezentowano analizę teoretyczną dynamiki ewoluujących populacji dla skrajnych liczebności populacji: nieskończonych i dwuelementowych, w jednowymiarowych rzeczywistych przestrzeniach poszukiwań bez ograniczeń. Wyniki teoretyczne uzyskano, jak do tej pory, tylko przy tych upraszczających założeniach, ponieważ metody ewolucyjne są skomplikowanymi nieliniowymi procesami stochastycznymi. Badano zarówno dynamikę lokalną - przy poszukiwaniu jedynego optimum jednomodalnych funkcji celu, jak i globalną - gdy funkcja celu była wielomodalna. Dla jasności wywodu w przypadku dynamiki globalnej ograniczono się głównie do funkcji dwumodalnych i analizy mechanizmu przejścia siodła pomiędzy optimami, aczkolwiek niektóre badania rozszerzono na funkcje o więcej niż dwóch optimach. Nieskończone populacje scharakteryzowano za pomocą rozkładów cech i opisano przez funkcje gęstości prawdopodobieństwa rozkładu osobników z daną cechą. Ustalono, że dla Instytut Informatyki, Automatyki i Robotyki Politechniki Wrocławskiej, ul. Janiszewskiego 11/17, 50-372 Wrocław. jednomodalnych funkcji celu populacja dąży do stanu równowagi wokół optimum, a jej rozkład jest normalny. Nieskończone populacje są więc zbieżne według średniej i z prawdopodobieństwem do optimum jednomodalnej funkcji celu. Dla funkcji wielomodalnych rozkład nieskończonych populacji jest mieszaniną rozkładów gaussowskich, których parametry określono. Rozkład sumaryczny mieszaniny jest przesunięty w stronę optimum globalnego. Jego wartość oczekiwana leży w pobliżu optimum globalnego, w odległości zależnej od wariancji mutacji. Symulacyjnie zbadano wpływ postaci funkcji celu oraz parametrów procesu ewolucyjnego na rozkłady populacji. Analizę dwuelementowych populacji przeprowadzono w przestrzeni stanów populacji, w której każdy punkt odpowiada całej populacji. Obliczono wartości oczekiwane stanów populacji, które generują dwuwymiarowy dyskretny układ dynamiczny. Głównym przedmiotem zainteresowań były trajektorie oraz zachowania asymptotyczne układu. Analizowano liczbę i położenie punktów stałych oraz określono warunki ich stabilności. Badano obszary przyciągania punktów stałych oraz szybkość zbieżności do nich. Liczba punktów stałych układu zależy od modalności funkcji celu, a także od wartości odchylenia standardowego mutacji. Punkty stałe są zwykle położone w pobliżu optimów funkcji celu i siodeł pomiędzy optimami. Udowodniono, że punkty stałe leżące w pobliżu siodeł są niestabilne. Stabilne punkty stałe położone w pobliżu optimów mogą stracić stabilność dla dużych wartości odchylenia standardowego mutacji. Pojawiają się wtedy orbity periodyczne. Dla niesymetrycznych funkcji celu obserwowano ciąg bifurkacji podwajania okresu prowadzący do zachowań chaotycznych. Zachowania chaotyczne badano następującymi metodami: wykresów fazowych, widma mocy, funkcji autokorelacji oraz wykładników Lapunowa. Uzyskane rezultaty badań dynamiki nieskończonych i dwuelementowych populacji wykorzystano do analizy innych aspektów działania metod ewolucyjnych. Badano wpływ różnych mechanizmów ewolucyjnych: selekcji, rekombinacji i mutacji, na różnorodność populacji. Wyniki symulacyjne potwierdziły wcześniejsze obserwacje dotyczące zmian różnorodności populacji w trakcie procesu ewolucyjnego. Początkowo znacznie rozproszone populacje szybko się skupiają, a początkowo jednorodne populacje zwiększają swą różnorodność. W efekcie, populacja w zaledwie kilku generacjach tworzy zwarty klaster osobników o zbliżonych typach. Największy wpływ na zwiększanie lub podtrzymywanie różnorodności populacji ma mutacja. Selekcja i w większości przypadków rekombinacja powodują zmniejszenie rozproszenia. Parametry krytyczne układu dynamicznego, dla których punkty stałe zmieniają stabilność, posłużyły do opracowania heurez dotyczących sposobów doboru parametrów algorytmu ewolucyjnego oraz identyfikacji nieznanej jednowymiarowej funkcji celu. W pracy dokonano także przeglądu znanych w literaturze, a związanych z tematem pracy, sposobów analizy metod ewolucyjnych: modele algorytmów fenotypowych oraz modele wykorzystujące łańcuchy Markowa i teorię układów dynamicznych. Przegląd uzupełniono krótkim opisem innych popularnych sposobów analizy metod ewolucyjnych.
The results of a theoretical analysis of an evolutionary search with the soft selection method are presented. The method is based on a simple model of Darwinian phenotypic evolution operating with proportional selection and Gaussian (normally distributed) mutation. In effect of superposition of both processes, populations could explore unbounded real-valued spaces, “searching” for better adaptations. Computer implementations of the method seem to be the key to the global optimisation in unbounded spaces of search. Especially, they have the ability to cross saddles between optima of a fitness function (thus capability of finding global optima), they display little sensitivity to an increase of the search space dimensionality and the robustness of inaccuracies and noise in a fitness function evaluation. The studied evolutionary process follows a Markov process with uncountable states, thus it is difficult to analyse it theoretically. In order to obtain theoretical results, some simplifying assumptions should be made. Two extremes of the population size are analysed: infinite and two-element ones. The search spaces are one-dimensional and without constraints. The local dynamics - in search of the only optimum of an unimodal fitness, and global dynamics - when a fitness function is multimodal, are studied. In the case of global dynamics, the studies are restricted to bimodal functions and the process of crossing saddle is a main issue. Then, some results are extended to fitness functions with more than two optima. Infinite populations are characterised by the distributions of traits and described by density of trait’s distribution functions. It is shown that for unimodal fitness functions, the distribution is normal and a population converges toward a selection- -mutation equilibrium state close to the optimum. For multimodal fitness functions, the distribution of population is a mixture of Gaussian distributions. Parameters of the mixture and component distributions are described. The influence of a fitness function topology and the parameters of evolutionary process on infinite population distribution is studied. Two-element populations are analysed in the space of population states, where each point corresponds to the whole population. The expected values of population states are calculated. These values generate a two-dimensional discrete dynamical system. Trajectories and asymptotic behaviour of the system are studied. The number and location of fixed points are analysed and the conditions of their stability are given. The number of fixed points depend on the modality of the fitness and on the standard deviation of mutation. Fixed points are usually located in the vicinity of 152 Dynamics of adaptation of phenotypic evolution methods optima and saddles of fitness functions. The basins of attraction of fixed points and the speed of convergence toward the points are studied. Fixed points being close to saddles are unstable. Stable fixed points close to optima can loose their stability at high values of the standard deviation of mutation and then periodical orbits appear. For an asymmetrical fitness, the sequences of double-period bifurcations leading to chaos are observed. Cyclic and chaotic behaviour is analysed using techniques for the dynamical systems’ theory: phase space portraits, auto-correlation function, power spectrum and Lyapunov exponents. The results obtained from the studies of the dynamics of infinite and two-elements populations are applied to investigate different aspects of evolutionary methods. The influence of various evolutionary operators on population diversity is studied. Simulation results confirmed our earlier observations that a population diversity does not change significantly during the evolutionary process. In initially widely distributed populations, the diversity is decreasing while in almost homogeneous populations, the diversity is increasing. As a result, a compact cluster of individuals with similar types drifts through a search space. Critical parameters derived from the dynamical system, for which fixed points change their stability, are used to set parameters of the evolutionary process and to identify unknown one-dimensional fitness functions. The survey of the known literature methods devoted to the theoretical analysis of evolutionary algorithms is also presented in the monograph. The models of phenotypic evolution, models exploiting the Markov chains, the theory of dynamical systems, schemes, etc., are described.
Czasopismo
Rocznik
Tom
Strony
156--156
Opis fizyczny
Bibliogr. 217 poz., rys., tab.
Twórcy
autor
Bibliografia
- [1] Altenberg L., The schema theorem and Price’s theorem, [w:] Proc. Foundations of Genetic Algorithms, pod red. D. Whitley, M. Vose, Morgan Kaufraann Publ., San Francisco 1995, p. 23-49.
- [2] Antonisse J., A new interpretation of schema notation that overturns the binary encodinq constraint, [w:l Conf. on Evolutionary Computation, Morgan Kaufmann Publ. 1989, p. 86-91.
- [3] Arabas J., Algorytmy ewolucyjne, WNT, Warszawa 2001.
- [4] Arabas J., Metoda ewolucyjna jako narzgdzie przeszukiwania przestrzeni Banacha, No. 152 [w:] Prace Naukowe, Oficyna Wyd. Polit. Warszawskiej, Warszawa 2005.
- [5] Ashlock W., Using very small population sizes in genetic programming, [w:] IEEE Congress on Evolutionary Computation, 2006, p. 319-326.
- [6] Ay tug H., Koehler G., New stopping criteria for genetic algorithms, European Journal of Operational Research 2000, Vol. 126, p. 662-674.
- [7] Baeck T,, Optimal mutation rates in genetic search, [w:] Int. Conf. on Genetic Algorithms, 1993, p. 2-8.
- [8] Baeck T., An overview of parameter control methods by self-adaptation in evolutionary algorithms, Fundamenta Informaticae 1998, Vol. 35, No. 1-4, p. 51-66.
- [9] Baeck T., Fogel D., Michalewicz Z., red., Handbook of evolutionary computation, Oxford Univ. Press, NY 1997.
- [10] Barker A., Martin W., Dynamics of a distance-based population diversity measure, [w:] Congress on Evolutionary Computation, Vol. 2, IEEE Press 2000, p. 1002-1009.
- [11] Barnsley M., Fractals Everywhere, Academic Press Inc. 1988. [12] Beyer H.G., Toward a theory of evolution strategies: on the benefit of sex, Evolutionary Computation 1995, Vol. 3, p. 81-111.
- [13] Beyer H.G., Toward a theory of evolution strategies: Self-adaptation, Evolutionary Computation 1996, Vol. 3, No. 3, p. 311-347.
- [14] Beyer H.G., The theory of evolution strategies, Springer-Verlag 2001.
- [15] Beyer H.G., Deb K., On self-adaptive features in real-parameter evolutionary algorithms, IEEE Trans. Evol. Comp. 2001, Vol. 5, No. 3, p. 250-270.
- [16] Bishop C., Neural networks for pattern recognition, Clarendon Press, Oxford 1995.
- [17] Boerlijst M.C., Bonhoeffer S., Nowak M.A., Viral quasi-species and recombination, Proc. Royal Society London B 1996, Vol. 263, p. 1577-1584.
- [18] Booker L., Fogel D., Whitley D., Angeline P., Recombination, [w:] Baeck i in. [9], p. C3.3.
- [19] Bose K., Kahng A., Muddu S., A new adaptive multistart technique for combinatorial global optimizations, Operations Research Letters 1994, Vol. 16, p. 101-113.
- [20] Branke J., Jin Y., red., Special issue on evolutionary computation in the presence of uncertainty, IEEE Trans. Evol. Comp. 2006, Vol. 10, No. 4. Bremermann H.J., Optimization through evolution and recombination, Spartan Books, Washington DC 1962.
- [21] Bruce D., Simpson R.J., Evolution determined by trajectory of expected populations: sufficient conditions, with application to crossover, Evolutionary Computation 1999, Vol. 7, No. 2, p. 151-171.
- [22] Bulier M.G., The mathematical theory of quantitative genetics, Clarendon Press, Oxford 1980.
- [23] Chakraborty U.K., Chakraborty M., Analysis of selection algorithms: a Markovchain approach, Evolutionary Computation 1996, Vol. 4, No. 2, p. 133-167.
- [24] Choręzyczewski A., Analiza wlasnosci adaptacyjnych ewoluujqcych populacji ze wzglgdu na zastosowania w optymalizacji globalnej, Rozprawa doktorska, Politechnika Wroclawska 2001.
- [25] Choręzyczewski A., Galar R., The impact of nonheritable variance on simulated evolutionary adaptation, [w:] KAEOiG, tagow 2003, p. 49-55.
- [26] Choręzyczewski A., Galar R., Adaptive saddle crossings in diverse landscapes, [w:] KAEOiG, Murzasichle 2006, p. 77-85.
- [27] Choręzyczewski A., Galar R., Karcz-Dulęba I., Considering phenotypic evolution in the space of population states, [w:] Proc. 5th Conf. Neural Networks and Soft Computing, pod red. R.T. L. Rutkowski, Zakopane, Poland 2000, p. 615-620.
- [28] Choręzyczewski A., Karcz-Dulęba I., On crossing saddles by small populations, [w:] KAEOiG, tagow 2003, p. 57-64.
- [29] Culberson J.C., Mutation-erossover isomorphism and the construction of discriminating functions, Evolutionary Computation 1995, Vol. 2, No. 3, p. 279-311.
- [30] Darwin C., On the origin of species by means of natural selection or the preservation of favoured races in the struggle for life, John Murray, London 1859.
- [31] Davis L., red., Handbook of Genetic Algorithms, Van Nostrand Reinhold, NY 1991.
- [32] Davis T.E., Principe J.C., A simulated like convergence theory for SGA, [w:] Int. Conf. on Genetic Algorithms, 1991, p. 174-181.
- [33] Davis T.E., Principe J.C., A Markov chain framework for the simple genetic algorithm, Evolutionary Computation 1993, Vol. 1, No. 3, p. 269-288.
- [34] De Jong K., Genetic algorithms are not function optimizers, [w:] Proc. Foundations of Genetic Algorithms, 1992, p. 5-17.
- [35] De Jong K., Spears W., Gordon D., Using Markov chains to analyze GAFOs, [w:] Proc. Foundations of Genetic Algorithms, pod red. L. Whitley, M. Vose, Vol. 3,
- [36] Morgan Kaufmann Publ. 1995, p. 115-137.
- [37] De Jong K.A., An analysis of the behavior of a class of genetic adaptive systems, Rozprawa doktorska, Univ. of Michigan, Ann Arbor, MI 1975.
- [38] Del Moral P., Miclo L., Patras F., Rubenthaler S., The convergence to equilibrium of neutral genetic models, ArXiv Mathematics e-prints 2007.
- [39] Devaney R.L., An introduction to chaotic dynamical systems, Addison-Wesley, Reading, M. A. 1989.
- [40] Droste S., Jansen T., Wegener I., Perhaps not a free lunch but at least a free appetizer, [w:] Genetic and Evolutionary Computation Conf, Orlando, FL 1999, p. 833-839.
- [41] Dua R., Hart P., Pattern classification and scene analysis, Wiley 1973.
- [42] Dulęba I., Karcz-Dulęba I., Zastosowania algorytmów genetycznych w robotyce, [w:] IV Krajowa Konferencja Robotyki, 1993, p. 243-250.
- [43] Dulęba I., Karcz-Dulęba I., Trajectory planning of robots: A GA approach, [w:] Cybemetics and Systems. Proc. 12 Europ. Meeting on Cybem. and Syst. Res., pod red. R. Trappl, Vol. 2, Vienna 1994, p. 1467-1474.
- [44] Dulęba I., Karcz-Dulęba I., Analiza dyskretnego układu dynamicznego generowanego przez pewien proces ewolucyjny, [w:] Proc. 9 Symp. Symulacja Procesów Dynamicznych, Zakopane 1996, p. 351-356.
- [45] Dulęba I., Karcz-Dulęba I., Algorithms of trajectory planning with constrained deuiation from a given end-effector path, Robotica 2004, Vol. 22, No. 6, p. 633-642.
- [46] Eiben A.E., Hinterding R., Michalewicz Z., Parameter control in euolutionary algorithms, IEEE Trans. Evol. Comp. 1999, Vol. 3, No. 2, p. 124-141.
- [47] Eigen M., Self-organization ofmatter and the euolution ofbiological macromolecules, Naturwissenschaften 1971, Vol. 58, p. 465-523.
- [48] Eigen M., Macromolecular euolution: dynamical ordering in sequence space, [w:] Emerging Synthesis in Science: Proc. Founding Workshops of the Santa Fe Institut, pod red. D. Pines, Addison Wesley, Reading MA 1985, p. 25-69.
- [49] Eigen M., Schuster P., The hypercycle - a principle ofnatural self-organization. Part A: emergency of the hypercycle, Naturwissenschaften 1977, Vol. 64, p. 541-565.
- [50] Eldredge N., Gould S., Punctuated equilibria: an altematiue to phylogenetic gradualism, [w:] Models in Paleobiology, pod red. T.J.M. Schopf, Freeman, Cooper and Co., San Francisco 1972, p. 82-115.
- [51] English T.M., On the structure of sequential search: Beyond “no free lunch”, [w:] Evolutionary Computation in Combinatorial Optimization - EuoCOP 2004, pod red. J. Gottlieb, G. Raidl, LNCS 3004, Springer-Verlag, Coimbra 2004, p. 95-103.
- [52] Eshelman L., Schaffer D., Real-coded genetic algorithms and interual schemata, [w:] Proc. Foundations of Genetic Algorithms, pod red. D.L. Whitley, Morgan Kaufmann Publ., San Mateo, CA 1993, p. 187-202.
- [53] Eshelman L.J., Schaffer D.J., Crossover’s niche, [w:] Int. Conf. on Genetic Algorithms, pod red. S. Forest, Morgan Kaufmann Publ. 1993, p. 9-14.
- [54] Feller W., Wstęp do rachunku prawdopodobieństwa, PWN, Warszawa 1981.
- [55] Fisher R.A., On the dominance ratio, Proceedings of the Royal Society of Edinburgh 1922, No. 42, p. 321-341.
- [56] Fogel D.B., An euolutionary approach to the traueling salesman problem, Biol. Cybern. 1988, Vol. 60, p. 139-144.
- [57] Fogel D.B., Evolving artificial intelligence, Rozprawa doktorska, Univ. of Całifornia, San Diego 1992.
- [58] Fogel D.B., Principles of euolutionary processes, [w:] Baeck i in. [9], p. A2.1.
- [59] Fogel D.B., Atmar J., Comparing genetic operators with Gaussian mutations in simulated euolutionary processes using linear systems, Biol. Cybern. 1990, Vol. 63, p. 111-114.
- [60] Fogel D.B., Stayton L., On the effectweness of crossouer in simulated euolutionary optimization, BioSystems 1994, Vol. 32, p. 171-182.
- [61] Fogel L., Owens A., Walsh M., Artificial Intelligence through Simulated Euolution, John Willey, New York 1966. Galar R., Handicapped individua in evolutionary processes, Biol. Cybern. 1985, Vol. 51, p. 1-9.
- [62] Galar R., Evolutionary search with soft selection, Biol. Cybern. 1989, Vol. 60, p. 357-364.
- [63] Galar R., Miękka selekcja w losowej adaptacji globalnej w Rn. Próba biocybernetycznego ujęcia rozwoju, Prace Naukowe Instytutu Cybernetyki Technicznej nr 84 Monografie nr 17, Wyd. Politechniki Wrocławskiej 1990.
- [64] Galar R., Knowledge economy and euolutionary traps, [w:] The knowledge based economy. The global challenges of21st century, pod red. A. Kukliński, W. Orłowski, Science and Gouemment Senes, Vol. 4, Rewasz, Warszawa 2000, p. 32-61.
- [65] Galar R., Knowledge management from the evolutionary perspective, [w:] Evolutionary computation and global optimization, pod red. J. Arabas, Oficyna Wyd. Polit. Warszawskiej 2002, p. 9-20.
- [66] Galar R., An adaptive approach to the regional strategy and its perception., [w:] Education and the futurę. Knowledge, education and territorial deuelopment in Central and Eastem European Countries, pod red. F. Russo, S. Arnaldi, Trieste, Italia 2003, p. 135-148.
- [67] Galar R., Freedom, perfection and adaptiue saddles, [w:] Intellectual entrepreneurship through or against institutions, pod red. S. Kwiatkowski, P. Houdayer, Wyd. Wyższej Szkoły Przeds. i Zarz. im. L. Koźmińskiego 2004, p. 53-70.
- [68] Galar R., Chorążyczewski A., Euolutionary dynamics in space of states, [w:] Congress on Euolutionary Computation, Seoul, Korea 2001, p. 1366-1373.
- [69] Galar R., Karcz-Dulęba I., The euolution of two: An example of space of states approach, [w:] Proc. 3rd Annual Conf. on Euolutionary Programming, pod red. A. Sebald, L. Fogel, World Scientific, San Diego CA 1994, p. 261-268.
- [70] Galar R., Karcz-Dulęba I., Najprostszy przypadek ewolucji: dynamika adaptacyjna dwuelementowych populacji, [w:] Systemy sterowania, pod red. W. Greblicki,
- [71] C. Smutnicki, WKŁ, Warszawa 2005, p. 65-76.
- [72] Galar R., Karcz-Dulęba I., Łojek R., Adaptation cost: what is the repayment time for the euolutionary saddle crossings?, [w:] KAEOiG, Będlewo 2007, p. 91-97.
- [73] Gallagher M., Freon M., Downs T., Real-valued euolutionary optimization using a flexible probability density estimator, [w:] Genetic and Euolutionary Computation Conf, Morgan, San Francisco, CA 1999, p. 840-846.
- [74] Geard N., Wiles J., Diversity maintenance on neutral landscapes: an argument for recombination, [w:] Congress on Euolutionary Computation, Vol. 1, Honolulu 2002, p. 211-213.
- [75] Goertzel B., A conuergence theoremfor the simple GA with population size tending to infinity, [w:] IEEE Int. Conf. on Euolutionary Computation, 1995, p. 74-79.
- [76] GoldbergD., Sizing populations for serial andparallel GA, [w:] Int. Conf. on Genetic Algońthms, Morgan Kaufmann Publ. 1989, p. 215-221.
- [77] Goldberg D.E., Genetic Algońthms in Search, Optimization, and Machinę Leaming, Addison-Wesley 1989.
- [78] Goldberg D.E., A note on Boltzmann toumament selection for genetic algorithms and population oriented simulated annealing, Complex Systems 1990, Vol. 4, p. 445-460.
- [79] Goldberg D.E., Algorytmy genetyczne w zastosowaniach, WNT, Warszawa 1995.
- [80] Goldberg D.E., Deb K., A comparatwe analysis of selection schemes used in genetic algońthms, [w:] Proc. Foundations of Genetic Algońthms, pod red. G. Rawlins, Morgan Kaufmann Publ., San Mateo, CA 1990, p. 69-93.
- [81] Goldberg D.E., Deb K., Clark J.H., Genetic algońthms, noise, and the sizing of populations, Complex Systems 1992, Vol. 6, No. 4, p. 333-362.
- [82] Goldberg D.E.,LingleR., Alleles, loci and the traveling salesmanproblem, [w:] ICGA and their applications, Lawrence Erlbaum, Mahwah, NJ 1985, p. 154-159.
- [83] Goldberg D.E., Segrest P., Finite Markov chain analysis of genetic algońthms, [w:] Genetic Algońthms and their Applications: Proc. Int. Conf. on Genetic Algońthms, pod red. J.J. Grefenstette, Lawrence Erlbaum, London 1987, p. 1-8.
- [84] Gould S.J., Eldredge N., Punctuated equilibńa: the tempo and mode of evolution reconsidered, Paleobiology 1977, Vol. 3, p. 115-251.
- [85] Grygiel K., On asymptotic properties of a selection-with-mutation operator, [w:] KAEOiG, Murzasichle 1996, p. 50-56.
- [86] Harik G.R., Cantu-Paz E., Miller B.L., Goldberg D.E., The gambler’s ruin problem, genetic algońthms, and the sizing of populations, [w:] IEEE Int. Conf. on Euolutionary Computation, IEEE Press, Indianapolis IN 1997, p. 7-12.
- [87] Hart W.E., Rethinking the design of real-coded euolutionary algońthms: Making discrete choices in continuous search domains, Soft Computing Journal 2005, Vol. 9, p. 225-235.
- [88] Haupt R.L., Haupt S.E., Optimum population size and mutation rate for a simple real genetic algońthm that optimizes anray factors, Applied Computational Electromagnetics Society Journal 2000, Vol. 15, No. 2, p. 1034-1037.
- [89] Herrera F., Lozano M., Verdegay J.L., Tackling real-coded genetic algońthms: Operators and toolsfor behauioural analysis, Artificial Intelligence Review 1998, Vol. 12, p. 265-319.
- [90] Hesser J., Manner R., Touiards an optimal mutation probability for genetic algońthms, [w:] Conf. on Parallel Problem Solving from Naturę, pod red. H.P. Schwefel, LNCS 496, Springer-Verlag 1991, p. 23-32.
- [91] Holland J.H., Adaptation in natural and artificial systems, Univ. of Michigan Press, Ann Arbor 1975.
- [92] Horn J., Finite Markov chain analysis of genetic algońthms with niching, [w:] Int. Conf. on Genetic Algońthms, pod red. S. Forrest, Morgan Kaufmann Publ., San Mateo, CA 1993, p. 110-117.
- [93] Horn J., Goldberg D., Genetic algońthm difficulty and the modality of fitness landscapes, [w:] Proc. Foundations of Genetic Algońthms, pod red. L. Whitley, M. Vose, Morgan Kaufmann Publ., San Francisco, CA 1995, p. 243-269.
- [94] Iosifescu M., Skończone procesy Markowa i ich zastosowania, PWN, Warszawa 1988. [95] Jones J., Evolutionary algońthms, fitness landscapes and search, Rozprawa doktorska, Univ. Of New Mexico 1995.
- [96] Karcz I., Badania symulacyjne stochastycznych procesów równowagi selekcyjno-mutacyjnej, Praca magisterska, Politechnika Wrocławska 1988.
- [97] Karcz-Dulęba I., Symulowane procesy ewolucyjne jako narzędzie optymalizacji globalnej w Rn, Rozprawa doktorska, Politechnika Wrocławska 1992.
- [98] Karcz-Dulęba I., Analiza wlaciwości adaptacyjnych dużych populacji, [w:] Sympozjum Symulacja procesów dynamicznych SPD-8, Polana Chochołowska 1994, p. 185-192.
- [99] Karcz-Dulęba I., Soft selection in d-optimal designs, [w:] Conf. on Parallel Problem Sohing from Naturę, pod red. Y. Davidor, et al, LNCS 866, Springer-Verlag, Jerusalem 1994, p. 608-616.
- [100] Karcz-Dulęba I., Niektóre aspekty zbieżności metody poszukiwań ewolucyjnych z miękką selekcją, [w:] KAEOiG, Rytro 1997, p. 113-119.
- [101] Karcz-Dulęba I., Dynamics of two element populations in the space of population states: the case of symmetrical cjuality functions, [w:] KAEOiG, Lądek Zdrój 2000, p. 115-122.
- [102] Karcz-Dulęba I., Dynamics of infinite populations evolving in a landscape ofuni- and bmodal fitness functions, IEEE Trans. Evol. Comp. 2001, Vol. 5, No. 4, p. 398-409.
- [103] Karcz-Dulęba I., Euolution of two-element population in the space of population states: eąuilibrium states for asymmetńcal fitness functions, [w:] KAEOiG, Jastrzębia Góra 2001, p. 106-113.
- [104] Karcz-Dulęba I., Domains of attraction of dynamical system generated by simple euolutionary model, [w:] Proc. Workshop on Genetic Algorithms, Bielsko-Biała - Szyndzielnia 2002, p. 20-22.
- [105] Karcz-Dulęba I., Euolution of two-element population in the space of population states: Eguilibrium states for asymmetrical fitness functions, [w:] Euolutionary Algorithms and Global Optimization, pod red. J. Arabas, Oficyna Wyd. Polit. Warszawskiej 2002, p. 35-46.
- [106] Karcz-Dulęba I., Asymptotic behauior of discrete dynamical system generated by simple euolutionary process, Journ. Appl. Math. and Comp. Sc. 2004, Vol. 14, No. 1, p. 79-90.
- [107] Karcz-Dulęba I., Concentration of population in phenotypic euolution, [w:] Artificial Intelligence and Soft Computing, pod red. L. Rutkowski, et al, LNAI 3070, Springer-Verlag 2004, p. 420-425.
- [108] Karcz-Dulęba I., Period-doubling bifurcations in discrete dynamical system generated by euolutionary process, [w:] KAEOiG, Kazimierz Dolny 2004, p. 83-88.
- [109] Karcz-Dulęba I., Time to the conuergence of euolution in the space of population states, Journ. Appl. Math. and Comp. Sc. 2004, Vol. 14, No. 3, p. 279-287.
- [110] Karcz-Dulęba I., Bifurkacje i chaos w ewolucji fenotypowej dla unimodalnych funkcji jakości, [w:] Systemy sterowania, pod red. W. Greblicki, C. Smutnicki, WKŁ, Warszawa 2005, p. 41-50.
- [111] Karcz-Dulęba I., Unstable behauior in a dynamical system generated by phenotypic euolution, [w:] Computer methods and systems, pod red. R. Tadeusiewicz, et al, Oprogramowanie Naukowo-Techniczne, Kraków 2005, p. 261-269.
- [112] Karcz-Dulęba I., Chaos detection by Lyapunou exponents in dynamical system generated by euolutionary process, [w:] Artificial Intelligence and Soft Computing, pod red. L. Rutkowski, LNCS, LNAI 4029, Springer-Verlag 2006, p. 380-389.
- [113] Karcz-Dulęba I., Dynamics of two-element populations in the space of population states, IEEE Trans. Evol. Comp. 2006, Vol. 10, No. 2, p. 199-209.
- [114] Karcz-Dulęba I., Identification of cyclic and chaotic behavior in a dynamical system generated by phenotypic evolution, [w:] KAEOiG, Kazimierz Dolny 2006, p. 195-202.
- [115] Karcz-Dulęba I., On use of unstable behavior of a dynamical system generated by phenotypic evolution, [w:] Knowledge-Driven Computing. Knowledge Engineering and Intelligent Computations, pod red. C. Cotta, et al, Studies in Computational Intelligence, Springer-Verlag 2008, p. 119-134.
- [116] Karcz-Dulęba I., Chorążyczewski A., Fraktalna metoda analizy granicznych rozkładów nieskończonych populacji, [w:] KAEOiG, Potok Złoty 1999, p. 141-148.
- [117] Karcz-Dulęba I., Dziemieszkiewicz P., Recombination and maintaining population diversity: a comparatwe study, [w:] KAEOiG, Korbielów 2005, p. 95-100.
- [118] Karcz-Dulęba I., Dziemieszkiewicz P., Recombination and maintaining population diversity: simulations results, [w:] KAEOiG, Korbielów 2005, p. 101-108.
- [119] Kimura M., A stochastic model conceming the maitenance of genetic variability in quantitative characters, Proc. Natl. Acad. USA 1965, Vol. 54, p. 731-736.
- [120] Kirkpatrick M., Quantum euolution and punctuated eguilibria in contimuous genetic characters, American Naturałist 1982, Vol. 119, No. 6, p. 833-848.
- [121] Kirkpatrick S., Gelatt C., Vecchi M., Optimization by simulated annealing, Science 1983, Vol. 220, No. 4598, p. 671-680.
- [122] Koehler G., A proof of the Vose-Liepins conjecture, Journal Annals of Mathematics and Artificial Intelligence 1994, Vol. 10, No. 4, p. 409-422.
- [123] Koehler G., Bhattacharyya S., Vose M., General cardinality genetic algońthms, Evolutionary Computation 1998, Vol. 5, No. 4, p. 439-459.
- [124] Kołodziej J., Modeling hierarchical genetic strategy as a Lindenmayer system, [w:] Proc. of Int. Conf. on Parallel Computing in Electńcal Engineeńng PARELEC ’2002, 2002, p. 409-415.
- [125] Kołodziej J., IIow accurately the infinite population model can approximate the behaviour of an euolutionary algońthm, [w:] KAEOiG, Kazimierz 2004, p. 89-97.
- [126] Kołodziej J., Ogiński L., The estimation ofthe population distribution by the mixture of Erlang distributions in euolutionary algońthms, [w:] KAEOiG, Korbielów 2005, p. 109-116.
- [127] Kwaśnicka H., Szerszon P., NetGen- evolutionary algońthms in designing artificial neural networks, [w:] Conf. Neural Networks and Their Applications, Kule 1997, p. 671-676.
- [128] Lande R., Natural selection and random genetic drift, Evolution 1976, Voł. 30, p. 314-334.
- [129] Leung Y., Gao Y., Xu Z., Degree of population diversity - a perspective onpremature convergence in Genetic Algońthms and its Markov chain analysis, IEEE Trans. Neural Networks 1996, Vol. 8, No. 5, p. 1165-1176.
- [130] Lindgren K., Euolutioriary phenomena in simple dynamics, [w:] Artificial Life II, pod red. C.G. Langton, et al, Addison-Wesley, Reading, MA 1992, p. 295-312.
- [131] Macdonald P., Analysis of length-frequency distributions, [w:] Age and Growth offish, pod red. R. Summerfelt, G. Hall, Iowa State Univ. Press, Ames 1987, p. 371-384.
- [132] Makuch M., Patan K., Obuchowicz A., Evolutionary search with soft selection in artificial neural network leaming process, [w:] KAEOiG, Murzasichle 1996, p. 87-93.
- [133] Mathias E., Whitley L., Genetic operators, the fitness landscape and the TSP, [w:J Conf. on Parallel Problem Solving from Nature, Elsevier 1992, p. 219-228.
- [134] Mattiussi K., Waibel M., Floreano D., Measures of diversity for populations and distances between individuals with highly reorganizable genomes, Evolutionary Computation 2004, Vol. 12, No. 4, p. 495-515.
- [135] Maynard Smith J., Evolutionary genetics, Oxford Univ. Press 1989.
- [136] McLachlan G.J., Peel D., Finite Mixture Models, Wiley, New York 2000.
- [137] Merz P., Freisleben B., Memetic algorithms and the fitness landscape of the graph bi-partitioning problem, [w:] Conf. on Parallel Problem Solving from Nature, Springer-Verlag, Berlin 1998, p. 765-774.
- [138] Michalewicz Z., A GA for the linear transportation problem, IEEE Trans. Syst. Man Cybernetics 1991, Vol. 21, No. 2, p. 445-452.
- [139] Michalewicz Z., Algorytmy genetyczne + struktury danych = programy ewolucyjne, WNT, Warszawa 1996.
- [140] Michalewicz Z., Fogel D.B., Jak to rozwiqzac, czyli nowoczesna heurystyka, WNT, Warszawa 2006.
- [141] Miroslaw L., Choręzyczewski A., Buchholz F., Kittler R., EA validation method in detection of mitotic cells, [w:] KAEOiG, Korbielow 2005, p. 157-162.
- [142] Moey C.C. J., Rowe J.E., A reduced Markov model ofGAs without the exact transition matrix, [w:] Conf. on Parallel Problem Solving from Nature, 2004.
- [143] Muhlenbein H., Towards a theory of organisms and evolving automata - open problems and ways to explore frontiers of evolutionary computation, [w:] Genetic Algorithms and Evolutionary Computation, pod red. A. Menon, Vol. 11, Kluwer Acad. Publ. 2004, p. 1-39.
- [144] Muhlenbein H., Mahnig T., Mathematical analysis of evolutionary algorithms for optimization, [w:] Proc. 3rd Int. Symposium on Adaptive Systems, pod red. A. Rodriguez, et al, Cuba 2001, p. 166-185.
- [145] Nimwegen E., Crutchfield J., Optimizing epochal evolutionary search: Population- -size dependent theory, Machine Learning 2001, Vol. 45, No. 1, p. 77-114.
- [146] Nimwegen E., Crutchfield J., Mitchell M., Finite populations induce metastability in evolutionary search, Physics Letters A 1997, Vol. 229, No. 3, p. 144-150.
- [147] Nimwegen E., Crutchfield J., Mitchell M., Statistical dynamics of the royal road genetic algorithm, Theor. Computer Science 1999, Vol. 214, p. 41-102.
- [148] Nix A.E., Vose M.D., Modeling genetic algorithms with Markov chains, Annals of Mathematics and Artificial Intelligence 1992, Vol. 5, p. 79-88.
- [149] Obuchowicz A., Evolutionary search with soft selection and deterioration of the objective function, [w:] Intelligent Information Systems, Symposium, pod red. M. Klopotek, et al, Wyd. PAN, Warszawa, Zakopane 1997, p. 288-295.
- [150] Obuchowicz A., Adaptation in a time-varying landscape using an evolutionary search with soft selection, [w:] KAEOiG, Wyd. Polit. Warszawskiej, Potok Zioty 1999, p. 245-251.
- [151] Obuchowicz A., Evolutionary search with soft selection in training a network of dynamic neurons, [w:] Intelligent Information Systems, Symposium, pod red. M. Klopotek, M. Michalewicz, Wyd. PAN, Warszawa, Ustron 1999, p. 214-223.
- [152] Obuchowicz A., Multi-dimensional Gaussian and Cauchy mutations, [w:] Intelligent Information Systems, pod red. M. Kłopotek, et al, Physica-Verlag, Heidelberg 2002, p. 133-142.
- [153] Obuchowicz A., Euolutionary algońthms for global optimization and dynamie system diagnosis, Lubusky Sc. Society, Zielona Góra 2003.
- [154] Ochoa G., Harley I., Buxton H., On recombination and optimal mutation rates, [w:] Genetic and Euolutionary Computation Conf, Morgan Kaufmann Publ. 1999, p. 13-17.
- [155] Ochoa G., Harvey I., Recombination and error thresholds in finite populations, [w:] Proc. Foundations of Genetic Algońthms, pod red. W. Banzhaf, C. Reeves, Morgan Kaufmann Publ. 1998, p. 245-264.
- [156] Peitgen H., Jurgens H., Saupe D., Granice chaosu. Fraktale, cz.l i ez. 2, PWN, Warszawa 1996.
- [157] Poli R., Langdon W., Schema theory for genetic programming with one-point crossouer and point mutation, Evolutionary Computation 1998, Vol. 6, No. 3, p. 231-252.
- [158] Poli R., McPhee N., Rowe J., Exact schema theory and Markou chain models for genetic programming and vańable-length genetic algońthms with homologous crossouer, Genetic Programming and Evolvable Machines 2004, Vol. 5, No. 1, p. 31-70.
- [159] Priigel-Bennett A., Modeling finite populations, [w:] Proc. Foundations of Genetic Algońthms, 2003, p. 99-114.
- [160] Priigel-Bennett A., Symmetry breaking in population-based optimization, IEEE Trans. Evol. Comp. 2004, Vol. 8, No. 1, p. 63-79.
- [161] Priigel-Bennett A., Rogers A., Modelling genetic algońthm dynamics, [w:] Theoretical Aspects of Euolutionary Computing, pod red. L. Kallel, et al, Springer-Verlag 2001, p. 59-86.
- [162] Priigel-Bennett A., Shapiro J., The dynamics of a genetic algońthm for simple random ising systems, Physica D 1997, Vol. 104, p. 75-114.
- [163] Qi X., Palmieri F., Theoretical analysis of evolutionary algońthms with an infinite population size in continuous space. Part I: Basic properties of selection and mutation, IEEE Trans. Neural Networks 1994, Vol. 5, p. 102-119.
- [164] Qi X., Palmieri F., Theoretical analysis of euolutionary algońthms with an infinite population size in continuous space. Part II: Analysis of the diversification role of crossouer, IEEE Trans. Neural Networks 1994, Vol. 5, p. 120-128.
- [165] R. M., Jong K.D., Measurement of population diuersity, [w:] Selected Papers from the 5th European Conference on Artificial Evolution, Springer-Verlag, London, UK 2002, p. 31-41.
- [166] Radcliffe N., Surry P., Fundamental limitations on search algońthms: Euolutionary computing in perspective., [w:] Computer Science Today, pod red. J. van Leeuwen, LNCS 1000, Springer-Verlag 1995, p. 275-291.
- [167] Radcliffe N.J., Forma analysis and random respectful recombination, [w:] Proc. Ąth Int. Conf. on Genetic Algońthms, pod red. R. Belew, L. Booker, Morgan Kaufmann Publ. 1991, p. 222-229.
- [168] Radcliffe N.J., Genetic set recombination, [w:] Proc. Foundations of Genetic Algońthms, pod red. D. Whitley, Morgan Kaufmann Publ., San Mateo, CA 1992, p. 203-219.
- [169] Rattray L.M., The dynamics of a genetic algońthm under stabilizing selection, Complex Systems 1995, Vol. 9, p. 213-234.
- [170] Rechenberg I., Evolutionsstrategie: Optimierung technischer Systeme nach Prinzipien der biologischen Erolution, Frohmann-Holzboog, Stuttgart 1973.
- [171] Rees J., Koehler G.J., An investigation of GA performance results for different cardinality alphabets, [w:] Evolutionary Algońthms, pod red. L.D. Davis, et al, Springer-Verlag, NY 1999, p. 191-206.
- [172] Reeves C.R., red., Modem heuństic techniques for combinatońalproblems, Blackwell Scien. Publ., Oxford, UK 1993.
- [173] Reeves C.R., Using GA with smali populations, [w:] Proc. 5 ICGA, Morgan Kaufmann Publ. 1993, p. 92-99.
- [174] Reeves C.R., Landscapes, operators and heuństic search, Annals of Operations Research 1999, Vol. 86, p. 473-490.
- [175] Reeves C.R., Rowe J., Genetic Algońthms: Pńnciples and perspectwes, Kluwer Acad. Publ. 2003.
- [176] Reeves C.R., Steele N., Problem soluing by simulated genetic process. a review and application to neural networks, [w:] Applied Informatics, lOth Int. Conf IASTED, 1992, p. 269-272.
- [177] Rocha M., Rocha I., Ferreira E.C., A new representation in evolutionary algońthms for the optimization of bioprocesses, [w:] IEEE Conf. on Euolutionary Computation, IEEE Press 2005, p. 484-490.
- [178] Rogers A., Priigel-Bennett A., The dynamics of a GA on a model hard, Complex Systems 2000, Vol. 11, No. 6, p. 437-464.
- [179] Rudolph G., Conuergence analysis ofcanonical genetic algońthm, IEEE Trans. Neural Networks 1994, Vol. 5, No. 1, p. 96-101.
- [180] Rudolph G., Conuergence properties of evolutionary algońthms, Verlag Dr. Kovac, Hamburg 1997.
- [181] Rudolph G., Stochastic processes, [w:] Baeck i in. [9], p. B2.2:1-8.
- [182] Rudolph G., Finite Markov chain results in evolutionary computation: A tour d’hońzon, Fundamenta Informaticae 1998, Vol. 35, No. 1-4, p. 67-89.
- [183] Schaefer R., Podstawy genetycznej optymalizacji globalnej, Wyd. UJ, Kraków 2002.
- [184] Schaefer R., Jabłoński Z.J., On the conuergence of sampling measures in the global genetic search, [w:] PPAM Conf., LNCS 2328, 2002, p. 593-600.
- [185] Schaefer R., Kołodziej J., Genetic serach reinforced by population hierarchy, [w:] Proc. Foundations of Genetic Algońthms, Morgan Kaufmann Publ. 2003, p. 383-401
- [186] Schaffer J., Caruna R., Eshelman L., Das R., A study of control parameters affecting online performance of genetic algońthms for function optimization, [w:] Int. Conf. on Genetic Algońthms, pod red. J. Schaffer, Morgan Kaufmann Publ. 1989, p. 51-60.
- [187] Schwefel H.P., Numerical optimization of Computer models, Wiley 1981.
- [188] Smith R., Population size, [w:] Baeck i in. [9], p. El.1:1-5.
- [189] Spears W., Crossover or mutation, [w:] Proc. Foundations of Genetic Algońthms, pod red. D. Whitley, Morgan Kaufmann Publ. 1992, p. 221-237.
- [190] Spears W.M., Evolutionary Algońthms: The Role of Mutation and Recombination., Springer-Verlag, Heidelberg 2000.
- [191] Spears W.M., JongK.A.D., Analysing GAsusingMarkov models withsematically ordered and lumped states, [w:] Proc. Foundations of Genetic Algorithms, pod red. R.K. Beelew, M.D. Vose, Morgan Kaufmann Publ., San Francisco, CA 1997, p. 85-100.
- [192] Strogatz S.H., Nonlinear Dynamics and Chaos, Addison-Wesley, Reading, MA 1994.
- [193] Suzuki H., Iwasa Y., Crossover accelerates evolution in gas with a babel-like fitness landscape: mathematical analyses, Evolutionary Computation 1999, Vol. 7, No. 3, p. 275-310.
- [194] Suzuki J., A Markou chain analysis on a genetic algorithm, [w:] Int. Conf. on Genetic Algorithms, 1993, p. 146-154.
- [195] Suzuki J., A Markou chain analysis on simple genetic algońthm, IEEE Trans. Syst. Man Cybernetics 1995, Vol. 25, No. 4, p. 655-659.
- [196] Syswerda G., A study of reproduction in generational and steady-state genetic algorithms, [w:] Proc. Foundations of Genetic Algońthms, Morgan Kaufmann Publ. 1991, p. 94-101.
- [197] Titterington D.M., Smith A.F., Makov U.E., Statistical analysis of finite mixture distńbutions, Wiley 1985.
- [198] Vazquez M., Whitley L.D., A compańson of genetic algońthm for the static job shop scheduling problem, [w:] Conf. on Parallel Problem Soluing from Naturę, 2000, p. 303-345.
- [199] Vitanyi P., Genetic fitness optimization using rapidly mixing Markou chains, [w:] 7th Int. Workshop on Algorithmic Leaming Theory, LNAI 1160, Springer-Verlag, Heidelberg 1996, p. 67-82.
- [200] Voget S., Theoretical analysis of GA with infinite population size, Complex Systems 1996, Vol. 10, No. 3, p. 167-183.
- [201] Vose M., Wright A.H., The simple genetic algońthm and the Walsh transform: Part /, theory, Evolutionary Computation 1998, Vol. 6, No. 3, p. 253-273.
- [202] Vose M., Wright A.H., The simple genetic algońthm and the Walsh transform: Part II, the inuerse, Evolutionary Computation 1998, Vol. 6, No. 3, p. 275-289.
- [203] Vose M.D., The Simple Genetic Algońthm. Foundations and Theory, MIT Press 1999.
- [204] Vose M.D., Liepins G.E., Punctuated equilibńa in genetic search, Complex Systems 1991, Vol. 5, p. 31-44.
- [205] Vose M.D., Wright A.H., Simple genetic algońthms with linear fitness, Evolutionary Computation 1994, Vol. 4, No. 2, p. 347-368.
- [206] Watson R.A., Pollack J.B., Recombination without respect: Schema combination and disruption in genetic algońthm crossouer, [w:] Genetic and Euolutionary Computation Conf, pod red. D. Whitley, et al, Morgan Kaufmann Publ. 2000, p. 112-119.
- [207] Wolpert A.H., Macready W.G., No free lunch theorems for optimization, IEEE Trans. Evol. Comp. 1997, Vol. 1, p. 67-82.
- [208] Wright A., Zhao Y., Markov chain models of genetic algońthms, [w:] Genetic and Evolutionary Computation Conf, Vol. 1, Morgan Kaufmann Publ., Orlando, Florida 1999, p. 734-741.
- [209] Wright A.H., Genetic algońthms for real parameter optimization, [w:] Proc. Foundations of Genetic Algońthms, pod red. G.J.E. Rawlins, Morgan Kaufmann Publ., San Mateo, CA 1991, p. 205-218.
- [210] Wright A.H., Agapie A., Cyclic and chaotic behavior in genetic algońthms, [w:] Proc. Genetic and Evolutionary Computation Conf, Morgan Kaufmann Publ., San Francisco 2001, p. 718-724.
- [211] Wright A.H., Bidwell G., A search for counterexamples to two conjectures on the simple genetic algońthm, [w:] Proc. Foundations of Genetic Algońthms, pod red. R.K. Belew, M.D. Vose, Morgan Kaufmann Publ., San Francisco, CA 1997, p. 73-84.
- [212] Wright A.H., Rowe J., Neil J., Analysis of the simple genetic algońthm on the single-peak and double-peak landscapes, [w:] Congress on Euolutionary Computation, IEEE Press 2002, p. 214-219.
- [213] Wright A.H., Rowe J., Stephens C., Poli R., Bistability in a gene pool GA with mutation, [w:] Proc. Foundations of Genetic Algońthms, Morgan Kaufmann Publ. 2003, p. 63-80.
- [214] Wright S., Euolution in Mandelian populations, Genetics 1931, Vol. 16, p. 97-159.
- [215] Wright S., Character change, speciation, and the higher taxa, Evolution 1982, Vol. 36, No. 3, p. 427- 443.
- [216] Yao X., Liu Y., Lin G., Evolutionary programming made faster, IEEE Trans. Evol. Comp. 1999, Vol. 3, No. 2, p. 82-102.
- [217] Zieliński R., Neumann P., Stochastyczne metody poszukiwania minimum funkcji, WNT, Warszawa 1986.
Typ dokumentu
Bibliografia
Identyfikator YADDA
bwmeta1.element.baztech-b7513562-f0a7-4d6e-8299-30c17f6b762c