Bin Yu

Timeline

1. 1978–1990

   Mathematics and a Berkeley doctorate

   After her family moves to Beijing in 1978, she returns to Harbin for a summer of mathematics lessons with her former teacher Jianye Chen. She earns a mathematics bachelor's degree at Peking University in 1984, followed by a Berkeley statistics master's in 1987 and PhD in 1990. At Berkeley, Lucien Le Cam and Terry Speed advise her doctoral research.

2. 1990–1993

   Wisconsin, Yale and Berkeley

   From 1990 to 1993, she is an assistant professor in Statistics at the University of Wisconsin–Madison, with a postdoctoral fellowship at Berkeley's Mathematical Science Research Institute in 1991 and a visiting assistant professorship in Statistics at Yale in 1993. She joins Berkeley's Department of Statistics as an assistant professor in 1993.

3. 1998–2006

   Bell Labs and Berkeley professorships

   From 1998 to 2000, she is a member of technical staff at Bell Labs, Lucent Technologies. She becomes a professor of Statistics at Berkeley in 2001 and a professor of Electrical Engineering and Computer Science in 2006. In 2006, she receives a Guggenheim fellowship to research interpretable models for high-dimensional data.

4. 2009–2014

   Statistics department and society leadership

   She chairs Berkeley's Department of Statistics from 2009 to 2012. From 2013 to 2014, she serves as president of the Institute of Mathematical Statistics.

5. 2013–2014

   Election to two academies

   In 2013, she is elected to the American Academy of Arts and Sciences. She is elected to the US National Academy of Sciences in 2014.

6. 2017–2018

   Iterative Random Forests

   With Sumanta Basu, Karl Kumbier and James B. Brown, she introduces iterative Random Forests (iRF) in a 2017 preprint, published in PNAS in 2018. The method trains an ensemble of decision trees to find stable interactions among multiple variables, with applications to gene regulation.

7. 2020

   The PCS framework

   In February 2020, she and Karl Kumbier publish Veridical Data Science in PNAS. Their predictability, computability and stability (PCS) framework checks predictions against reality and assesses how choices in data cleaning, modeling and interpretation affect results.

8. 2022–2023

   Deep learning at Microsoft Research

   From 2022 to 2023, she works as a half-time consultant researcher in the deep learning group at Microsoft Research in Redmond.

9. 2024

   The Veridical Data Science book

   In 2024, she and Rebecca L. Barter publish the online book Veridical Data Science: The Practice of Responsible Data Analysis and Decision Making. The book uses PCS stress tests throughout a data science project, from defining the problem and cleaning data to modeling and communicating results.

10. As of October 2026

    Trustworthy AI at Berkeley

    As of October 2026, she is Chancellor's Distinguished Professor in Berkeley's Departments of Statistics and Electrical Engineering and Computer Sciences, with research on trustworthy AI, interpretability and deep learning theory. In June 2026, she leads Green Shielding with UCSF's Kornblith Group to evaluate language models for medical diagnosis, and co-leads a veridical AI theme at Norway's TRUST centre with Arnoldo Frigessi.