
Acceleration of Symbolic Regression Using Cartesian Genetic Programming
Hodaň, David ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
This thesis is focused on finding procedures that would accelerate symbolic regressions in Cartesian Genetic Programming. It describes Cartesian Genetic Programming and its use in the task of symbolic regression. It deals with the SIMD architecture and the SSE and AVX instruction set. Several optimizations that lead to a significant acceleration of evolution in Cartesian Genetic Programming are presented. A method of a bitlevel parallel simulation that uses AVX2 vectors allows to process 256 input combinations of a logic circuit in paralell. Similarly it is possible to use a bytelevel parallel simulation and work with 32 bytes when evolving an image filter. A new method of batch mutation can accelerate the evolution of combinational logic circuits thousand times depending on the problem size. For example, using a combination of these and other methods the evolution of 5 x 5b multipliers took 5.8 seconds on average on an Intel Core i54590 processor.


Deep Neural Networks Approximation
Stodůlka, Martin ; Mrázek, Vojtěch (referee) ; Vaverka, Filip (advisor)
The goal of this work is to find out the impact of approximated computing on accuracy of deep neural network, specifically neural networks for image classification. A version of framework Caffe called Ristrettocaffe was chosen for neural network implementation, which was extended for the use of approximated operations. Approximated computing was used for multiplication in forward pass for convolution. Approximated components from Evoapproxlib were chosen for this work.


Application of SAT Solvers in Circuit Optimization Problem
Minařík, Vojtěch ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
This thesis is focused on the task of application of SAT problem and it's modifications in area of evolution logic circuit development. This task is supposed to increase speed of evaluating candidate circuits by fitness function in cases where simulation usage fails. Usage of SAT and #SAT problems make evolution of complex circuits with high input number significantly faster. Implemented solution is based on #SAT problem. Two applications were implemented. They differ by the approach to checking outputs of circuit for wrong values. Time complexity of implemented algorithm depends on logical complexity of circuit, because it uses logical formulas and it's satisfiability to evaluate logic circuits.


Exploiting Approximate Arithmetic Circuits in Neural Networks Inference
Matula, Tomáš ; Mrázek, Vojtěch (referee) ; Češka, Milan (advisor)
Táto práca sa zaoberá využitím aproximovaných obvodov v neurónových sieťach so zámerom prínosu energetických úspor. K tejto téme už existujú štúdie, avšak väčšina z nich bola príliš špecifická k aplikácii alebo bola demonštrovaná v malom rozsahu. Pre dodatočné preskúmanie možností sme preto skrz netriviálne modifikácie opensource frameworku TensorFlow vytvorili platformu umožňujúcu simulovať používanie approximovaných obvodov na populárnych a robustných neurónových sieťach ako Inception alebo MobileNet. Bodom záujmu bolo nahradenie väčšiny výpočtovo náročných častí konvolučných neurónových sietí, ktorými sú konkrétne operácie násobenia v konvolučnách vrstvách. Experimentálne sme ukázali a porovnávali rozličné varianty a aj napriek tomu, že sme postupovali bez preučenia siete sa nám podarilo získať zaujímavé výsledky. Napríklad pri architektúre Inception v4 sme získali takmer 8% úspor, pričom nedošlo k žiadnemu poklesu presnosti. Táto úspora vie rozhodne nájsť uplatnenie v mobilných zariadeniach alebo pri veľkých neurónových sieťach s enormnými výpočtovými nárokmi.


Monitoring of Temperature for Small Buildings
Handzuš, Jakub ; Wiglasz, Michal (referee) ; Mrázek, Vojtěch (advisor)
The aim of this thesis is to design and implement an IoT system for monitoring the air temperature of smaller objects, e.g households. As the system is to be financially available to the wider public, it needs to be fully functional even at low procurement and operating costs  based on this requirement, it is necessary to analyze available alternatives for operating the systems. When selecting the appropriate technology for storing the acquired data, it is necessary to take into account the operations most frequently performed on the data  for this reason, a set of experiments is carried out on several types of database systems. Based on the findings gathered during analysis and experimentation, the optimal solution appears to be the combination of a generic database with web host services. In the resulting system, the sensor sends the acquired data to the server with a database, whilst the processed data is subsequently interpreted by the clientside visualizations.


Autonomous Impedance Meter
Voda, Zbyšek ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
This thesis deals with design of a smart embedded device for autonomous measurement of impedance optimized for measurement of biological materials. The goal is to create a device which provides a simple web interface which allows users to capture and further analyze measured data. The digital part is based on MT7688 SoC with WiFi capabilities. The analogue part utilizes a singlechip integrated solution AD5933 that is tightly coupled with a custom analogue frontend whose function is to modify the signal to avoid a potential damage of biological samples. The proposed device is able to either measure impedance for a single frequency or perform a frequency sweep across the whole range beginning at 50Hz and ending at 100 kHz. It supports common twoelectrode probes as well as more precise fourelectrode probes. The analogue frontend has been simulated using the SPICE simulator to avoid a potential design bug. The experimental evaluation shows that the achieved precision for the typical impedance of biological samples is better than 0.5%.


A Library for Convolutional Neural Network Design
Rek, Petr ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
In this diploma thesis, the reader is introduced to artificial neural networks and convolutional neural networks. Based on that, the design and implementation of a new library for convolutional neural networks is described. The library is then evaluated on widely used datasets and compared to other publicly available libraries. The added benefit of the library, that makes it unique, is its independence on data types. Each layer may contain up to three independent data types  for weights, for inference and for training. For the purpose of evaluating this feature, a data type with fixed point representation is also part of the library. The effects of this representation on trained net accuracy are put to a test.


Deep Neural Networks: Embedded System Implementation
Matěj, Aleš ; Šimek, Václav (referee) ; Mrázek, Vojtěch (advisor)
The goal of this thesis is to firstly design and implement an application for embeddedsystems which will classify MNIST numbers and secondly optimize energy and memoryrequirements of this network. The basics of neural networks, CortexM processor cores andembedded devices are described in the theoretical part. Followed by implementation details.Networks learning is done with Python and Theano library on a PC. The network is thenconverted to C for a board STM32F429 Discovery. Last part consist of network optimization,which focuses on convolution, dot product and number representation of weights and biasesof the network.


Evolutionary Optimization of Convolutional Neural Networks
Roreček, Pavel ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
This Master's Thesis is focused on the principles of neural networks, primarily convolutional neural networks (CNN). It introduces the evolutionary optimization in the context of neural networks. One of existing libraries devoted to the CNN design was chosen (Keras), analysed and used in image classification tasks. An optimization technique based on cartesian genetic programming that should reduce the complexity of CNN's computation was proposed and implemented. The impact of the proposed technique on CNN behaviour was evaluated in a case study.


Analysis of the Israeli farright activities on the West Bank
Mrázek, Vojtěch ; Jelen, Libor (advisor) ; Nováček, Aleš (referee)
There are many negative consequences related to the Israeli occupation of the West Bank. The phenomenon of the settler's violence is one of them. The aim of the thesis is to explain the conditionality of the violence and to put it into context with Israeli political scene. A quantitative analysis is made to examine the relationship between Israeli farright parties' electoral gains and the incidence of violence against Palestinians residents of the West Bank. Also, spatial analysis is made to measure the level of clustering of the incidents. To articulate the theoretical assumptions, the theory of social cleavages was used. In line with the assumptions, the results suggest that in Jewish settlements on the West Bank, there is statistically significant connection between Israeli farright parties' electoral gains and the number of the violent incidents. The strongest correlation is proven between the incidents and the electoral gains of the parties influenced by Kahanism, a militant racist ideology. There is a weaker correlation between the incidence of violence and the electoral gains of the parties representing Religious Zionism, an ideology that combines religion and nationalism. The spatial distribution of the incidents was irregular. There was a clustering of high values in several areas,...
