Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Explore research on evading machine learning-based malware detection in this 13-minute conference talk that demonstrates how attackers can use ML techniques to defeat ML security systems. Learn how minor modifications to malware can significantly reduce detectability by ML-based security tools, while examining the broader implications of deploying machine learning security solutions without proper scrutiny. Discover the vulnerabilities inherent in ML security systems and understand why organizations must carefully evaluate these tools before implementation. Gain insights into the adversarial relationship between offensive and defensive machine learning applications in cybersecurity, presented through practical examples of evasion techniques that highlight the limitations of current ML-based detection methods.
