search menu icon-carat-right cmu-wordmark

Human-Computer Decision Systems Poster (SEI 2015 Research Review)

Describes work to use learning theory advances to account for persistent human expert teams and experiments to improve the human-computer decision systems.

Software Engineering Institute


Security decision systems aim to distinguish malicious activity from benign and often use a combination of human experts and automated analysis, including machine learning (ML). Systems using only human experts scale poorly; pure ML systems are susceptible to structured attacks by adversaries and, in most cases, have unsatisfactory performance on their own.