Dead-end Discovery: How offline reinforcement learning could assist healthcare decision-makers

Опубликовано 2 февраля 2022, 19:47

Microsoft Research Senior Researcher Mehdi Fatemi, MIT Assistant Professor Marzyeh Ghassemi, and PhD student Taylor W. Killian answer several questions about their NeurIPS 2021 paper, “Medical Dead-ends and Learning to Identify High-risk States and Treatments.” They discuss how their research in offline reinforcement learning and their AI model could alert a human decision-maker when a set of decisions are potentially high-risk, such as in healthcare scenarios.

Read the blog post here: microsoft.com/en-us/research/b...

Read the full paper here: microsoft.com/en-us/research/p...