Towards Automated Design of Complex Modular Systems Inspired by Nature

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00236676" target="_blank" >RIV/68407700:21230/15:00236676 - isvavai.cz</a>

Výsledek na webu

<a href="http://cyber.felk.cvut.cz/teaching/radaUIB/disertace%20Vitku%20Jaroslav.pdf" target="_blank" >http://cyber.felk.cvut.cz/teaching/radaUIB/disertace%20Vitku%20Jaroslav.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Towards Automated Design of Complex Modular Systems Inspired by Nature

Popis výsledku v původním jazyce

Research in these days is composed of small, isolated and often highly specialized sub- fields. One of main goals of the thesis is to contribute to simplification of reusing of outcomes of research of others. This is done by enabling the combination of various pieces of software together into a bigger system. The thesis proposes a framework inspired in modular artificial neural networks. The framework represents various sub-systems (pieces of code) in an unified way. Together with unified communicationprotocol between these subsystems (called Neural Modules), the framework enables arbitrary combinations of these Modules into bigger, modular sys- tems. A complexity of each Module can range from the simplest mathematical operation towards the complicated artificial intelligence algorithms. Here, the simulator capable of general-purpose modular systems is proposed. Together with the simulator, the resulting framework enables user to design and test (by means of rapid prototyping) complex

Název v anglickém jazyce

Towards Automated Design of Complex Modular Systems Inspired by Nature

Popis výsledku anglicky

Research in these days is composed of small, isolated and often highly specialized sub- fields. One of main goals of the thesis is to contribute to simplification of reusing of outcomes of research of others. This is done by enabling the combination of various pieces of software together into a bigger system. The thesis proposes a framework inspired in modular artificial neural networks. The framework represents various sub-systems (pieces of code) in an unified way. Together with unified communicationprotocol between these subsystems (called Neural Modules), the framework enables arbitrary combinations of these Modules into bigger, modular sys- tems. A complexity of each Module can range from the simplest mathematical operation towards the complicated artificial intelligence algorithms. Here, the simulator capable of general-purpose modular systems is proposed. Together with the simulator, the resulting framework enables user to design and test (by means of rapid prototyping) complex

Druh

O - Ostatní výsledky

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů