معرفی
Finale Doshi-Velez is the Herchel Smith Professor of Computer Science at the Harvard John A. Paulson School of Engineering and Applied Sciences. She holds a MSc from the University of Cambridge (as a Marshall Scholar) and a PhD from MIT, followed by a postdoc at Harvard Medical School. Her research focuses on probabilistic methods for human-AI decision-making, spanning healthcare applications, AI interpretability, and socio-technical AI policy. Her DtAK group develops tools for accountable AI systems, including Bayesian models, decision-making frameworks, and validation techniques.
Key research areas include:
- Probabilistic Modeling: Uncertainty quantification in heterogeneous data, Bayesian inference, and model reliability.
- Decision-Making: Reinforcement learning for clinical policies, inverse reinforcement learning, and sequential decision support.
- Interpretability: Explanations for AI policies, concept bottleneck models, and human-AI collaboration.
Notable awards include the Sloan Fellowship, AFOSR YIP, NSF CAREER Award, and the Everett Mendelsohn Excellence in Mentoring Award. Her work bridges technical innovation with ethical AI governance, including contributions to regulatable AI systems and procurement checklists. She also explores creative writing, with fantasy novels centering South Asian perspectives.
