Skip to main content

Research Repository

Advanced Search

Feature vulnerability and robustness assessment against adversarial machine learning attacks (2021)
Conference Proceeding
Mccarthy, A., Andriotis, P., Ghadafi, E., & Legg, P. (in press). Feature vulnerability and robustness assessment against adversarial machine learning attacks

Whilst machine learning has been widely adopted for various domains, it is important to consider how such techniques may be susceptible to malicious users through adver-sarial attacks. Given a trained classifier, a malicious attack may attempt to cra... Read More about Feature vulnerability and robustness assessment against adversarial machine learning attacks.

Shouting through letterboxes: A study on attack susceptibility of voice assistants (2020)
Presentation / Conference
Mccarthy, A., Gaster, B., & Legg, P. (2020, June). Shouting through letterboxes: A study on attack susceptibility of voice assistants. Paper presented at IEEE International Conference on Cyber Security and the Protection of Digital Services (Cyber Science 2020)

Voice assistants such as Amazon Echo and Google Home have become increasingly popular for many home users, for home automation, entertainment, and convenience. These devices process speech commands from a user to execute some action, such as playing... Read More about Shouting through letterboxes: A study on attack susceptibility of voice assistants.