Bc. Michal Savčinský

Master's thesis

Využití strojového učení pro tvorbu optimálních útočných strategií

Using machine learning to optimize attack strategies
Abstract:
V súčasnosti dosahuje strojové učenie významné úspechy v mnohých oblastiach, ako autonómne riadenie áut, virtuálni asitenti alebo hranie hier. Z toho dôvodu veríme, že schopnosti strojového učenia by mohli byť použité aj v oblasti kyberbezpečnosti. Na vyvinutie robustného a odolného systému voči sofistikovanému správaniu útočníka potrebujeme náš obranný systém natrénovať proti širokému spektru útočných …more
Abstract:
Recently, reinforcement learning methods have accomplished significant breakthroughs in many areas, such as autonomous driving, virtual assistants, or games. Therefore we believe that the capabilities of reinforcement learning could be exploited for cybersecurity as well. To develop a robust system, proof against sophisticated adversary behavior, we need to train our defensive system against the broad …more
 

Keywords

machine learning reinforcement learning neural network recurrent neural network long short-term memory Q-learning
 
Language used: English
Date on which the thesis was submitted / produced: 20. 5. 2019
Identifier: https://is.muni.cz/th/zkmth/

Thesis defence

  • Date of defence: 21. 6. 2019
  • Supervisor: RNDr. Martin Drašar, Ph.D.
  • Reader: RNDr. Tomáš Jirsík

Citation record

Cite this text

Full text of thesis

Contents of on-line thesis archive
Published in Theses:
  • světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatiky
Složka Reference to the local database directory of the institution