Humans vs. Machines in Malware Classification

Explore a groundbreaking experimental study comparing human and machine intelligence in malware classification through this 15-minute conference talk from USENIX Security '23. Discover how 110 human players worldwide, including both novices and experts, approached malware classification based on sandbox reports. Learn about the surprising similarities in feature prioritization between experts and novices, and compare their performance to state-of-the-art Machine Learning models. Gain insights into the differences in decision-making processes and feature extraction between humans and computer algorithms. Understand the implications of these findings for training malware analysts and improving feature encoding in cybersecurity.

USENIX Security '23 - Humans vs. Machines in Malware Classification