Website (URL)
Scholar Link
Interests
Biography
Miruna Oprescu is a SUNY PRODiG+ Fellow in the Department of Computer Science at Stony Brook University, with a planned tenure-track transition in 2028 following successful review. She earned her Ph.D. in Computer Science at Cornell University, advised by Nathan Kallus. Her doctoral work was supported by a DOE Computational Science Graduate Fellowship. Previously, she was a Senior Data and Applied Scientist at Microsoft Research, where she made foundational contributions to EconML, an open-source Python library for causal machine learning. Oprescu holds an A.B. in Physics and Mathematics from Harvard University.
Research
Oprescu is building the Trustworthy Inference, Learning, Decision-making, and Evaluation (TILDE) Lab. She develops principled machine learning methods for settings where data are imperfect, assumptions are uncertain, and actions shape future observations. Her interests span three connected themes:
- Inference: What can we reliably learn from imperfect evidence? Causal inference, statistical machine learning, and uncertainty quantification, including learning from observational, experimental, and structured data.
- Learning and decision-making: How should an agent act and learn through interaction? Reinforcement learning, adaptive experimentation, and policy evaluation, with broader interests in learning through interaction and human feedback.
- Trustworthiness: When can we trust the resulting conclusions and behavior? Robustness, reliable evaluation, and understanding the limits of what data can establish.
Her research connects these methodological questions with applications in healthcare, climate and Earth systems, neuroscience, and energy and industrial systems.