Data science underpins modern AI and many advances in healthcare, yet human judgment permeates every stage of the data science life cycle. These judgment calls introduce hidden uncertainties that go well beyond sampling variability and drive many of the risks associated with AI.
At this year's Milne Lecture, come learn how Professor Bin Yu and her research group are addressing those uncertainties to help improve the way AI is leveraged to heal patients.
Date: Monday, May 11, 2026
Time: 4 to 5 p.m. (a reception will precede the lecture from 3:30 to 4 p.m.)
Location: Johnson Hall 102
Bin Yu, professor of statistics at UC Berkeley, will introduce veridical data science: Grounded in three fundamental principles — Predictability, Computability and Stability (PCS) — veridical data science makes the uncertainties surrounding human judgment more explicit and assessable, aggregating reality-checked algorithms for better results. The PCS framework unifies and extends best practices in statistics and machine learning and is illustrated through healthcare applications, including identifying genetic drivers of heart disease, reducing cost of prostate cancer detection, improving uncertainty quantification beyond standard conformal prediction, and proposing Green Shielding, a new user-centric framework for safeguarding users of AI.
Be sure to arrive early for a reception from 3:30 to 4 p.m. right outside Johnson 102!
Professor Bin Yu leads the Yu Group at UC Berkeley, an interdisciplinary team in Statistics and EECS dedicated to advancing machine learning, artificial intelligence and veridical data science. Her group develops efficient and interpretable ML/AI methods and theory, ranging from iterative random forests (iRF) and tree-based FIGS to LoRA+ for fine-tuning deep learning and CD-T and SPEX for interpreting deep models. The Yu Group collaborates closely with domain experts in medical AI, genomics and neuroscience.
A member of both the National Academy of Sciences and the American Academy of Arts and Sciences, she has delivered major keynotes, including the 2019 Breiman Lecture at NeurIPS, 2023 IMS Wald Lectures and the COPSS DAAL Lecture (formerly Fisher Lecture). She and her team pioneered the PCS framework (predictability, computability, and stability) for veridical (truthful) data science (VDS), which has become an influential guide for transparent, trustworthy data science and AI.
