Secure Machine Learning

Independent study investigating the impact of adversarial attacks including data poisoning and model evasion on ML model reliability and performance.