- Добавил: literator
- Дата: 11-03-2023, 03:57
- Комментариев: 0
Название: Cyber Deception: Techniques, Strategies, and Human Aspects
Автор: Tiffany Bao, Milind Tambe, Cliff Wang
Издательство: Springer
Серия: Advances in Information Security
Год: 2023
Страниц: 252
Язык: английский
Формат: pdf (true)
Размер: 10.2 MB
This book introduces recent research results for cyber deception, a promising field for proactive cyber defense. The beauty and challenge of cyber deception is that it is an interdisciplinary research field requiring study from techniques and strategies to human aspects. This book covers a wide variety of cyber deception research, including game theory, Artificial Intelligence, cognitive science, and deception-related technology. We seek to create deceptive behaviors by leveraging evasion attacks against deep neural networks (DNNs). In particular, we propose to model an attacker as a DNN whose input is a trace of the observable behavior of a defended system. We then attempt evasion attacks that modify the observed behavior of the defended system such that the modified behavior obeys the above constraints: deceiving the attacker (into taking some action other than the action that would compromise the defended system), while remaining compatible with the original intended behavior of the system.
Автор: Tiffany Bao, Milind Tambe, Cliff Wang
Издательство: Springer
Серия: Advances in Information Security
Год: 2023
Страниц: 252
Язык: английский
Формат: pdf (true)
Размер: 10.2 MB
This book introduces recent research results for cyber deception, a promising field for proactive cyber defense. The beauty and challenge of cyber deception is that it is an interdisciplinary research field requiring study from techniques and strategies to human aspects. This book covers a wide variety of cyber deception research, including game theory, Artificial Intelligence, cognitive science, and deception-related technology. We seek to create deceptive behaviors by leveraging evasion attacks against deep neural networks (DNNs). In particular, we propose to model an attacker as a DNN whose input is a trace of the observable behavior of a defended system. We then attempt evasion attacks that modify the observed behavior of the defended system such that the modified behavior obeys the above constraints: deceiving the attacker (into taking some action other than the action that would compromise the defended system), while remaining compatible with the original intended behavior of the system.