معرفی
Michael Oberst is an Assistant Professor of Computer Science at Johns Hopkins University's Whiting School of Engineering, affiliated with the Malone Center for Engineering in Healthcare and the Data Science and AI Institute. His research focuses on developing reliable machine learning systems for healthcare decision-making, emphasizing causal inference and robust performance across diverse clinical settings.
Key research themes include:
- Ensuring ML system reliability comparable to FDA-approved medical tools
- Causal reasoning in observational healthcare data
- Robustness to dataset shifts across hospitals
- Algorithmic fairness under unobserved confounding
- Medical adaptation of large language models
Recent publications (2025-2024) demonstrate trends in prediction-powered inference, clinical validation frameworks, and robustness evaluation methods. His work appears in top ML venues (NeurIPS, ICML, UAI, EMNLP) and translational medicine journals.
Michael holds a BS in Statistics from Harvard University and a PhD in Computer Science from MIT, with postdoctoral training at Carnegie Mellon University's Machine Learning Department. His group actively seeks PhD students and postdocs for developing trustworthy AI solutions in healthcare.
