|
|
Instruction-Controlled Cellular Automata
Bendl, Jaroslav ; Žaloudek, Luděk (referee) ; Bidlo, Michal (advisor)
The thesis focuses on a new concept of cellular automata control based on instructions. The instruction can be understood as a rule that checks the states of cells in pre-defined areas in the cellular neighbourhood. If a given condition is satisfied, the state of the central cell is changed according to the definition of the instruction. Because it's possible to perform more instructions in one computational step, their sequence can be understood as a form of a short program. This concept can be extended with simple operations applied to the instruction's prescription during interpretation of the instructions - an example of such operation can be row shift or column shift. An advantage of the instruction-based approach lies in the search space reduction. In comparison with the table-based approach, it isn't necessary to search all the possible configurations of the cellular neighbouhood, but only several areas determined by the instructions. While the groups of the inspected cells in the cellular neighbourhood are designed manually on the basis of the analysis of the solved task, their sequence in the chromosome is optimized by genetic algorithm. The capability of the proposed method of cellular automata control is studied on these benchmark tasks - majority, synchronization, self-organization and the design of combinational circuits.
|
|
|
Evolutionary Circuit Design at the Transistor Level
Žaloudek, Luděk ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
This project deals with evolutionary design of electronic circuits with an emphasis on digital circuits. It describes the theoretical basics for the evolutionary design of circuits on computer systems, including the explanation of Genetic Programming and Evolutionary Strategies, possible design levels of electronic circuits, CMOS technology overview, also the overview of the most important evolutionary circuits design methods like development and Cartesian Genetic Programming. Next introduced is a new method of digital circuits design on the transistor level, which is based on CGP. Also a design system using this new method is introduced. Finally, the experiments performed with this system are described and evaluated.
|
|
| |
|
|
ALPS Technique in Cartesian Genetic Programming
Stanovský, Peter ; Slaný, Karel (referee) ; Sekanina, Lukáš (advisor)
This work introduces a brief summary of softcomputing and the solutions to NP-hard problems. It especially deals with evolution algorithms and their basic types. The next part involves the study of cartesian genetic programming, which belongs to the field of evolution algorithms, used mainly in the evolution of digital circuits, symbolic regression, etc. A special chapter is devoted to the studies of new technique Age layered population structure, which deals with the problems of premature convergence, which suggests the way of how the population could be divided into subpopulations split up according to the age criteria. Thanks to the maintaining of sufficient diversity, it achieves substantially better solutions in comparison to the classical evolution algorithms. This papier includes the suggestion of two ways of incorporation of the ALPS technique into CGP. In the next part of work there were carried out tests on the classic problems, that would be solved with evolution algorithms. These tests were made with and without using ALPS technique. In the part of work "Experimental results" there was discussed a contribution of using ALPS technique in CGP against the classic CGP.
|
|
|
A Tool for Visual Analysis of Circuit Evolution
Staurovská, Jana ; Minařík, Miloš (referee) ; Sekanina, Lukáš (advisor)
The main goal of the master's thesis is to compose a study on cartesian genetic programming with focus on evolution of circuits and to design a concept for visualisation of this evolution. Another goal is to create a program to visualise the circuit evolution in cartesian genetic programming, its generations and chromosomes. The program is capable of visualising the changes between generations and chromosomes and comparing more chromosomes at once. Several user cases had been prepared for the resulting program.
|
|
|
Coevolutionary Algorithm in FPGA
Hrbáček, Radek ; Vašíček, Zdeněk (referee) ; Drahošová, Michaela (advisor)
This thesis deals with the design of a hardware acceleration unit for digital image filter design using coevolutionary algorithms. The first part introduces reconfigurable logic device technology that the acceleration unit is based on. The theoretical part also briefly characterizes evolutionary and coevolutionary algorithms, their principles and applications. Traditional image filter designs are compared with the biologically inspired design methods. The hardware unit presented in this thesis exploits dual MicroBlaze system extended by custom peripherals to accelerate cartesian genetic programming. The coevolutionary image filter design is accelerated up to 58 times. The hardware platform functionality in the task of impulse noise filter design and edge detector design has been empirically analyzed.
|
|
|
Prediction of Secondary Structure of Proteins Using Cellular Automata
Brigant, Vladimír ; Drahošová, Michaela (referee) ; Bendl, Jaroslav (advisor)
This work describes a method of the secondary structure prediction of proteins based on cellular automaton (CA) model - CASSP. Optimal model and CA transition rule parameters are acquired by evolutionary algorithm. Prediction model uses only statistical characteristics of amino acids, so its prediction is fast. Achieved results was compared with results of other tools for this purpose. Prediction cooperation with a existing tool PSIPRED was also tested. It didn't succeed to beat this existing tool, but partial improvement was achieved in prediction of only alpha-helix secondary structure motif, what can be helful if we need the best prediction of alpha-helices. It was developed also a web interface of designed system.
|
|
|
Learnable Evolution Model for Optimization (LEM)
Grunt, Pavel ; Vašíček, Zdeněk (referee) ; Schwarz, Josef (advisor)
My thesis is dealing with the Learnable Evolution Model (LEM), a new evolutionary method of optimization, which employs a classification algorithm. The optimization process is guided by a characteristics of differences between groups of high and low performance solutions in the population. In this thesis I introduce new variants of LEM using classification algorithm AdaBoost or SVM. The qualities of proposed LEM variants were validated in a series of experiments in static and dynamic enviroment. The results have shown that the metod has better results with smaller group sizes. When compared to the Estimation of Distribution Algorithm, the LEM variants achieve comparable or better values faster. However, the LEM variant which combined the AdaBoost approach with the SVM approach had the best overall performance.
|
|
|
Cellular Automata Design Optimization
Jílek, Tomáš ; Drábek, Vladimír (referee) ; Bidlo, Michal (advisor)
Focus of this master's thesis is on evolutionary design of cellular automata and it's optimalization. There are described Evolutionary algorithms and Cellular automata in first part. Thereafter, one of the new ways of transition function representation and its possible evolutionary design is presented. Name of this method is Conditionally Matching Rules. This is followed by description of realized experiments with detailed results. Finally, success with optimalization for some tasks is presented in last chapter along with discussion.
|
|
|
Colearning in Coevolutionary Algorithms
Wiglasz, Michal ; Dobai, Roland (referee) ; Drahošová, Michaela (advisor)
Cartesian genetic programming (CGP) is a form of genetic programming where candidate programs are represented in the form of directed acyclic graphs. It was shown that CGP can be accelerated using coevolution with a population of fitness predictors which are used to estimate the quality of candidate solutions. The major disadvantage of the coevolutionary approach is the necessity of performing many time-consuming experiments to determine the best size of the fitness predictor for the particular task. This project introduces a new fitness predictor representation with phenotype plasticity, based on the principles of colearning in evolutionary algorithms. Phenotype plasticity allows to derive various phenotypes from the same genotype. This allows to adapt the size of the predictors to the current state of the evolution and difficulty of the solved problem. The proposed algorithm was implemented in the C language and optimized using SSE2 and AVX2 vector instructions. The experimental results show that the resulting image filters are comparable with standard CGP in terms of filtering quality. The average speedup is 8.6 compared to standard CGP. The speed is comparable to standard coevolutionary CGP but it is not necessary to experimentally determine the best size of the fitness predictor while applying coevolution to a new, unknown task.
|
guest ::
