Adversarial Attack’s Impact on Machine Learning Model in Cyber-Physical Systems
Abstract:
Deficiency of correctly implemented and robust defence leaves Internet of Things devices vulnerable to cyber threats, such as adversarial attacks. A perpetrator can utilize adversarial examples when attacking Machine Learning models used in a cloud data platform service. Adversarial examples are malicious inputs to ML-models that provide erroneous model outputs while appearing to be unmodified. This kind of attack can fool the classifier and can prevent ML-models from generalizing well and from learning high-level representation; instead, the ML-model learns superficial dataset regularity. This study focuses on investigating, detecting, and preventing adversarial attacks towards a cloud data platform in the cyber-physical context.
AUTHORS
Faculty of Information Technology University of Jyväskylä, Jyväskylä
Finland
Petri Vähäkainu is a researcher (MSc. MIT, MSc. (econ.), BSc. (econ.), B. Eng.) in the Finnish Defence Research Agency FDRA and cybersecurity doctoral student in Faculty of Information Technology at the University Jyväskylä, Finland. He has been researching utilization of Artificial Intelligence in cybersecurity, data science, health care, and Structural Health Monitoring.
University of Jyväskylä, Jyväskylä
Finland
Dr. Martti Lehto, PhD (Military Sciences), Col (GS) (ret.) works as a cybersecurity professor in the University of Jyväskylä. He has over 30 years of experience in C5ISR Systems in Finnish Defence Forces. He is also adjunct professor at the National Defence University in Air and Cyber Warfare. He has over 200 publications in the areas of C5ISR systems, cyber security and defence, information warfare, artificial intelligence, air power, and defence policy.
Faculty of Information Technology University of Jyväskylä, Jyväskylä
Finland
Antti Kariluoto researches artificial intelligence and its applicability at the University of Jyväskylä. He is an avid artificial intelligence enthusiast with a passion for data science.
Published In
Journal of Information Warfare
The definitive publication for the best and latest research and analysis on information warfare, information operations, and cyber crime. Available in traditional hard copy or online.
Quick Links
Archive
