Researchers detail TrojAI, a framework for hardening AI models against adversarial attacks
March 17, 2020
One way to test machine learning models for robustness is with what’s called a trojan attack, which involves modifying a model to respond to input triggers that cause it to infer an incorrect response. In an attempt to make these tests more repeatable and scalable, researchers at Johns Hopkins University developed a framework dubbed TrojAI, a set of tools that generate triggered data sets and…