Parameter Based Attack

Parameter-based attacks target the internal parameters of machine learning models to degrade their performance or extract sensitive information. Research currently focuses on improving the effectiveness of these attacks, particularly against embedded systems and in scenarios with limited data manipulation, exploring vulnerabilities in various model architectures including convolutional and fully-connected networks. Understanding and mitigating these attacks is crucial for ensuring the security and privacy of machine learning systems deployed in diverse applications, driving ongoing efforts to develop robust evaluation methodologies and effective defenses.