|
|
Automated compression of neural network weights
Lorinc, Marián ; Sekanina, Lukáš (referee) ; Mrázek, Vojtěch (advisor)
Konvolučné neurónové siete (CNN) od svojho vynájdenia zrevolucionizovali spôsob, akým sa realizujú úlohy z odvetvia počítačového videnia. Vynález CNN viedol k zníženiu pamäťovej náročnosti, keďže váhy boli nahradené konvolučnými filtrami obsahujúcimi menej trénovateľných váh. Avšak, toto zníženie bolo dosiahnuté na úkor zvýšenia požiadaviek na výpočtový výkon, ktorý je naviazaný na výpočet konvolúcie. Táto práca skúma hypotézu, či je možné sa vyhnúť načítavaniu váh a miesto toho ich vypočítať, čím sa ušetrí energia. Na otestovanie tejto hypotézy bol vyvinutý nový algoritmus kompresie váh využívajúci Kartézske genetické programovanie. Tento algoritmus hľadá najoptimálnejšiu funkciu kompresie váh s cieľom zvýšiť energetickú účinnosť. Experimenty vykonané na architektúrach LeNet-5 a MobileNetV2 ukázali, že algoritmus dokáže efektívne znížiť spotrebu energie pri zachovaní vysokej presnosti modelu. Výsledky ukázali, že určité vrstvy je možné doplniť vypočítanými váhami, čo potvrdzuje potenciál pre energeticky efektívne neurónové siete.
|
|
|
Polymorphic Self-Checking Circuits
Mazuch, Martin ; Růžička, Richard (referee) ; Sekanina, Lukáš (advisor)
This Master's thesis deals with question of the development of self-checking polymorphic circuits. It deals with a traditional way of creating reliable and self-checking circuits, presenting basic principles and methods. Also a method of Cartesian Genetic Programming for development of combinational circuits is explained. This thesis describes concepts of polymorphic gates and circuits and their benefits in practical use. Some existing self-checking polymorphic circuits are presented and their self-checking capabilities are analyzed. A proposal of realization of a design system for self-checking polymorphic circuits is given. A design system has been built based on presented specification and an application allowing simulations and analysis of system-proposed solutions has been created. Variety of experiments have been performed at created system and several interesting solutions have been acquired. At the end, conclusion is given and benefits of MSc. project are discussed.
|
|
|
Approximation of Digital Circuits in Yosys Tool
Plevač, Lukáš ; Vašíček, Zdeněk (referee) ; Mrázek, Vojtěch (advisor)
The goal of this work is introduction of cgploss extension. This extension is extension for combinational logic circuits optimization in Yosys tool. Cartesian genetic programming is introduced in the first part of this work. Cartesian genetic programming is a design and optimization method that can be used for circuit optimization and approximation. This chapter introduces representation of combinational logic circuits for Cartesian genetic programming. The next chapter introduces Yosys tool and possibilities of the Yosys extending. The proposed 'cgploss' extension is introduced in the next chapter. The chapter also provides details about the implementation and the usage. The last chapter tests cgploss extension and compares representation of combinational logic circuits.
|
|
|
Application of Approximate Computing in Image Processing
Hruda, Petr ; Vašíček, Zdeněk (referee) ; Bidlo, Michal (advisor)
This master thesis focuses on approximate computing applied to image processing. Specifically, the approximation is applied to adaptive thresholding. Two approaches were used, the design of a new system using approximated components and the approximation of an existing algorithm. The resulting effect on thresholding quality was investigated. Experimental evaluation of the first approach shows quality improvements of thresholding with usage of aproximated components. Also, area of found aproximated solutions is smaller. Evaluation of the second approach shows worse quality of thresholding with usage of aproximated components. The second approach is then declared inappropriate.
|
|
|
Cartesian Genetic Programming in Evolutionary Art
Veselý, Pavel ; Hyrš, Martin (referee) ; Petrlík, Jiří (advisor)
This thesis deals with use of Cartesian Genetic Programming (CGP) in Evolutionary Art (EvoArt). Text presents introduction to the topic. The rest of the thesis focuses on the process of design, implementation and testing of new method of application CGP in EvoArt. The proposed method uses CGP to create 2D vector images. Web application for EvoArt creation is made to demonstrate this method. Achieved results are presented and evaluated.
|
|
|
Problem Encoding Methods in Evolutionary Design of Combinational Circuits
Sedláček, Adam ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
The thesis compares two different approaches to combinational circuit encoding for automated circuit design which uses evolutionary algorithms. The comparison was made between cartesian genetic programming and circuit represented in the algebraic normal form. Both methods were evaluated on a chosen set of circuits. The first test case criterion was the convergence of each particular method. The second optimization criterion was the area used on a chip. For accelerating the evaluation of fitness a parallel simulation was used. Implementation is in programming language C++ with Boost library. The pros and cons of both methods are summarised at the end of this work.
|
|
|
Approximation of Digital Circuits in Yosys Tool
Plevač, Lukáš ; Vašíček, Zdeněk (referee) ; Mrázek, Vojtěch (advisor)
The goal of this work is introduction of cgploss extension. This extension is extension for combinational logic circuits optimization in Yosys tool. Cartesian genetic programming is introduced in the first part of this work. Cartesian genetic programming is a design and optimization method that can be used for circuit optimization and approximation. This chapter introduces representation of combinational logic circuits for Cartesian genetic programming. The next chapter introduces Yosys tool and possibilities of the Yosys extending. The proposed 'cgploss' extension is introduced in the next chapter. The chapter also provides details about the implementation and the usage. The last chapter tests cgploss extension and compares representation of combinational logic circuits.
|
|
|
Problem Encoding Methods in Evolutionary Design of Combinational Circuits
Sedláček, Adam ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
The thesis compares two different approaches to combinational circuit encoding for automated circuit design which uses evolutionary algorithms. The comparison was made between cartesian genetic programming and circuit represented in the algebraic normal form. Both methods were evaluated on a chosen set of circuits. The first test case criterion was the convergence of each particular method. The second optimization criterion was the area used on a chip. For accelerating the evaluation of fitness a parallel simulation was used. Implementation is in programming language C++ with Boost library. The pros and cons of both methods are summarised at the end of this work.
|
|
|
Application of Approximate Computing in Image Processing
Hruda, Petr ; Vašíček, Zdeněk (referee) ; Bidlo, Michal (advisor)
This master thesis focuses on approximate computing applied to image processing. Specifically, the approximation is applied to adaptive thresholding. Two approaches were used, the design of a new system using approximated components and the approximation of an existing algorithm. The resulting effect on thresholding quality was investigated. Experimental evaluation of the first approach shows quality improvements of thresholding with usage of aproximated components. Also, area of found aproximated solutions is smaller. Evaluation of the second approach shows worse quality of thresholding with usage of aproximated components. The second approach is then declared inappropriate.
|
|
|
Cartesian Genetic Programming in Evolutionary Art
Veselý, Pavel ; Hyrš, Martin (referee) ; Petrlík, Jiří (advisor)
This thesis deals with use of Cartesian Genetic Programming (CGP) in Evolutionary Art (EvoArt). Text presents introduction to the topic. The rest of the thesis focuses on the process of design, implementation and testing of new method of application CGP in EvoArt. The proposed method uses CGP to create 2D vector images. Web application for EvoArt creation is made to demonstrate this method. Achieved results are presented and evaluated.
|
guest ::
