MIT: Machine Learning Models Too Judgmental on Rule Violations

CAMBRIDGE, Mass., May 10, 2023 - In an effort to improve fairness or reduce backlogs, machine learning models are sometimes designed to mimic human decision making, such as deciding whether social media posts violate toxic content policies. But researchers from MIT and elsewhere have found that these models often do not replicate human decisions about [...]

The post MIT: Machine Learning Models Too Judgmental on Rule Violations appeared first on High-Performance Computing News Analysis | insideHPC.