Adversarial Machine Learning: The Maritime Context

Adversarial machine learning (ML) is the study of the capabilities of attackers and their goals, as well as the design of attack methods that exploit the vulnerabilities of ML during the ML life cycle. Since the discovery of the adversarial vulnerabilities of deep neural networks by Google Deepmind in 2014, the field of adversarial ML has been experiencing exponential growth. Last year, the Australian Signals Directorate published “Guidelines for secure AI system development”, which include precautions against adversarial ML, highlighting the impact of adversarial ML. Together with the Australian Institute for Machine Learning, we have been investigating adversarial ML in the space context. I will talk about the work we have done in the space context, and extend the lessons learnt to the maritime context. I will also discuss what other adversarial ML researchers have done in the maritime context, and highlight some implications that maybe of interest to the maritime community.

Event Venue & Nearby Stays

Engineering South Lecture Room S112, University of Adelide, 259 North Terrace, Adelaide SA 5000, 259 North Terrace, Adelaide, Australia