Browsing by Subject "evolutionary algorithms"

Now showing 1 - 15 of 15

Access status: Restricted ,
Algorytmy genetyczne + struktury danych = programy ewolucyjne
(Wydawnictwa Naukowo-Techniczne, 2003) Michalewicz, Zbigniew
Access status: Restricted ,
Badanie dynamiki populacji agentów autonomicznych w zmiennym środowisku w zależności od postaci praw ewolucji
(Data obrony: 2017-12-01) Wawrzos, Marek
Wydział Elektrotechniki, Automatyki, Informatyki i Inżynierii Biomedycznej
Access status: Open Access ,
Efficient selection methods in evolutionary algorithms
(Wydawnictwa AGH, 2024) Stańczak, Jarosław
Evolutionary algorithms mimic some elements of the theory of evolution. The survival of individuals and the ability to produce offspring play significant roles in the process of natural evolution. This process is called natural selection. This mechanism is responsible for eliminating weaker members of the population and provides the opportunity for the development of stronger individuals. The evolutionary algorithm, an instance of evolution in the computer environment, also requires a selection method – a computerized version of natural selection. Widely used standard selection methods applied in evolutionary algorithms are usually derived from nature and prefer competition, randomness, and some kind of “fight” among individuals. But the computer environment is quite different from nature. Computer populations of individuals are typically small, making them susceptible to premature convergence towards local extremes. To mitigate this drawback, computer selection methods must incorporate features distinct from those of natural selection. In the computer selection methods randomness, fight, and competition should be controlled or influenced to operate to the desired extent. This work proposes several new methods of individual selection, including various forms of mixed selection, interval selection, and taboo selection. The advantages of incorporating them into the evolutionary algorithm are also demonstrated, using examples based on searching for the maximum ?-clique problem and traditional Traveling Salesman Problem (TSP) in comparison with traditionally considered highly efficient tournament selection, deemed ineffective proportional (roulette) selection, and other classical methods.
Access status: Restricted ,
Efficient storage of training data for DNN (pytorch-ddnevo) evolution platform
(Data obrony: 2020-09-29) Szeremeta, Grzegorz
Wydział Informatyki, Elektroniki i Telekomunikacji
Access status: Restricted ,
Evolutionary techniques of investment portfolio construction
(Data obrony: 2014-10-29) Kocot, Mikołaj
Wydział Informatyki, Elektroniki i Telekomunikacji
Access status: Open Access ,
Metoda deterioracji funkcji celu dla algorytmów poszukiwań ewolucyjnych z miękką selekcją
(Wydawnictwa AGH, 2009) Orzechowski, Patryk
In this paper we present an initial version of fitness function deterioration method by applying linear combination of Gaussian functions. The algorithm may be used as a part of evolutionary search algorithm with soft selection. Experiments show that algorithm is especially helpful for population to cross saddles of multi-modal functions. Further research is needed to optimally assign the algorithm parameters, as well as to set its use cases.
Access status: Open Access ,
Multi-objective optimization of vehicle routing problem using evolutionary algorithm with memory
(Wydawnictwa AGH, 2017) Podlaski, Krzysztof; Wiatrowski, Grzegorz
The idea of a new evolutionary algorithm with memory aspect included is proposed to find multiobjective optimized solution of vehicle routing problem with time windows. This algorithm uses population of agents that individually search for optimal solutions. The agent memory incorporates the process of learning from the experience of each individual agent as well as from the experience of the population. This algorithm uses crossover operation to define agents evolution. In the paper we choose as a base the Best Cost Route Crossover (BCRC) operator. This operator is well suited for VPRTW problems. However it does not treat both of parent symmetrically what is not natural for general evolutionary processes. The part of the paper is devoted to find an extension of the BCRC operator in order to improve inheritance of chromosomes from both of parents. Thus, the proposed evolutionary algorithm is implemented with use of two crossover operators: BCRC and its extended-modified version. We analyze the results obtained from both versions applied to Solomon’s and Gehring & Homberger instances. We conclude that the proposed method with modified version of BCRC operator gives statistically better results than those obtained using original BCRC. It seems that evolutionary algorithm with memory and modification of Best Cost Route Crossover Operator lead to very promising results when compared to the ones presented in the literature.
Access status: Open Access ,
Niching in evolutionary multi-agent systems
(Wydawnictwa AGH, 2013) Krzywicki, Daniel
Niching is a group of techniques used in evolutionary algorithms, useful in several types of problems, including multimodal or nonstationary optimization. This paper investigates the applicability of these methods to evolutionary multi-agent systems (EMAS), a hybrid model combining the advantages of evolutionary algorithms and multi-agent systems. This could increase the efficiency of this type of algorithms and allow to apply them to a wider class of problems. As a starting point, a simple but flexible EMAS framework is proposed. Then, it is shown how to extend this framework in order to introduce niching, by adapting two classical niching methods. Finally, preliminary experimental results show the efficiency and the simultaneous discovery of multiple optima by this modified EMAS.
Access status: Open Access ,
Optimization of material distribution for forged automotive components using hybrid optimization techniques
(Wydawnictwa AGH, 2021) Sebastjan, Przemysław; Kuś, Wacław
The paper deals with the problem of optimal material distribution inside the provided design area. Optimization based on deterministic and stochastic algorithms is used to obtain the best result on the basis of the proposed objective function and constraints. The optimization of the shock absorber is used as an example of the described methods. One of the main difficulties addressed is the manufacturability of the optimized part intended for the forging process. Additionally, nonlinear buckling simulation with the use of the finite element method is used to solve the misuse case of shock absorber compression, where the shape of the optimized part has a key role in the total strength of the automotive damper. All of that, together with the required design precision, creates the nontrivial constrained optimization problem solved using the parametric, implicit geometry representation and a combination of stochastic and deterministic algorithms used with parallel design processing. Two methods of optimization are examined and compared in terms of the total amount of function calls, final design mass, and feasibility of the resultant design. Also, the amount of parameters used for the implicit geometry representation is greatly reduced compared to existing schemes presented in the literature. The problem addressed in this article is strongly inspired by the actual industrial example of the mass minimization process, but it is more focused on the actual manufacturability of the resultant component and admissible solving time. Commercially accessible software combined with authors' procedures is used to resolve the material distribution task, which makes the proposed method universal and easily adapted to other fields of the optimization of mechanical elements.
Access status: Open Access ,
Optymalizacja parametrów algorytmu ewolucyjnego w zadaniu regulacji napięcia w rozległej sieci rozdzielczej
(2008) Gruszka, Krzysztof
Treścią artykułu jest problem doboru parametrów algorytmu ewolucyjnego, stosowanego do optymalizacji regulacji napięcia w rozległych sieciach rozdzielczych. W artykule przedstawiono metodę poszukiwań optymalnych wartości parametrów, mających decydujący wpływ na efektywność algorytmu, oraz zamieszczono wybrane wyniki badań.
Access status: Open Access ,
Regulacja procesu spalania z wykorzystaniem SSN oraz algorytmów ewolucyjnych
(Wydawnictwa AGH, 2008) Kalita, Mariusz; Wójcik, Waldemar
The paper presents simulation of applying evolutionary algorithms and neural networks in power boiler process regulation. The various methods of selection, niche mechanism and coupling methods are presented in the paper. The results show that implementing proper selection operator and niche mechanism allow to prevent premature convergence and reduce computation time.
Access status: Open Access ,
The advent of the deep learning evolutionary algorithm EvoDN2 and its recent applications
(Wydawnictwa AGH, 2025) Chakraborti, Nirupam
The evolutionary deep learning algorithm EvoDN2 is an emerging strategy for data-driven intelligent learning and many-objective optimisation capable of handling a large volume of noisy and non-linear data. This article provides the essential details of this algorithm and highlights a number of its recent applications.
Access status: Open Access ,
Warunkowa wartość oczekiwana funkcji celu w konstrukcji algorytmów przybliżonych dla zagadnień permutacyjnych
(Wydawnictwa AGH, 2005) Chmiel, Wojciech; Kadłuczka, Piotr
W artykule zaprezentowano rezultaty prac nad zaawansowanymi algorytmami ewolucyjnymi stosowanymi w optymalizacji zagadnień permutacyjnych. Wprowadzenie dodatkowego parametru - warunkowej wartości oczekiwanej funkcji celu pozwala na ocenę jakości rozwiązań częściowo ustalonych. Może być on stosowany jako pomocnicze kryterium wyboru i konstrukcji nowych rozwiązań, w celu poprawy efektywności projektowanych algorytmów.
Access status: Restricted ,
Wykłady z algorytmów ewolucyjnych
(Wydawnictwa Naukowo-Techniczne, 2004) Arabas, Jarosław
Access status: Open Access ,
Wykorzystanie własności zagadnienia QAP w konstrukcji algorytmów ewolucyjnych
(Wydawnictwa AGH, 2007) Kadłuczka, Piotr; Chmiel, Wojciech
The paper presents an approach to an implementation and evaluation of evolutionary algorithm using operators exploiting peculiar properties of QAP problem. They are based on expected conditional value of objective function for partially fixed solutions. The numerical experiments were performed for standard test problems of quadratic assignment problem (QAP) from QAPLIB-A library. We compare the results of algorithms using pseudo-genetic operators which exploit some QAP problem properties with results obtained from algorithms using standard pseudo-genetic operators for permutation problems.