Bin Yu
Chancellor's Distinguished Professor and Class of 1936 Second Chair
Departments of Statistics and Electrical Engineering and Computer Sciences
Chan-Zuckerberg Biohub Investigator Alumnus • Weill Neurohub Investigator
mail: 367 Evans Hall #3860 • Berkeley, CA 94720
phone: 510-642-2781 • fax: 510-642-7892 • binyu@berkeley.edu
Welcome
I'm Bin Yu, the head of the Yu Group at Berkeley, which consists of 15-20 students and postdocs from Statistics and EECS. I was formally trained as a statistician, but my research interests and achievements extend beyond the realm of statistics. Together with my group, my work has leveraged new computational developments to solve important scientific problems by combining novel statistical machine learning approaches with the domain expertise of my many collaborators in neuroscience, genomics and precision medicine. We also develop relevant theory to understand random forests and deep learning for insight into and guidance for practice.
We have developed the PCS framework for veridical data science (or responsible, reliable, and transparent data analysis and decision-making). PCS stands for predictability, computability and stability, and it unifies, streamlines, and expands on ideas and best practices of machine learning and statistics.
In order to augment empirical evidence for decision-making, we are investigating statistical machine learning methods/algorithms (and associated statistical inference problems) such as dictionary learning, non-negative matrix factorization (NMF), EM and deep learning (CNNs and LSTMs), and heterogeneous effect estimation in randomized experiments (X-learner). Our recent algorithms include staNMF for unsupervised learning, iterative Random Forests (iRF) and signed iRF (s-iRF) for discovering predictive and stable high-order interactions in supervised learning, contextual decomposition (CD) and aggregated contextual decomposition (ACD) for interpretation of Deep Neural Networks (DNNs).
My vision for data science - papers & talks
"What is uncertainty in today's practice of data science?" (Yu, 2023). Journal of Econometrics.
"Building trust in medical AI algorithms with veridical data science" (interview of Bin Yu by Dr. Merle Behr for the German scientific journal KI - Künstliche Intelligenz), 2023
Veridical data science (PCS framework: v-flow code and documentation template), PNAS, 2020 (QnAs with Bin Yu)
Breiman Lecture (video) at NeurIPS "Veridical data Science" (PCS framework and iterative random forests (iRF)), 2019; updated slides, 2020
Stability, Bernoulli, 2013
Stability expanded, in reality, Harvard Data Science Review (HDSR), 2020.
Data science process: one culture. JASA, 2020.
Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist, Nature Medicine, 2020.
Definitions, methods and applications in interpretable machine learning, PNAS, 2019
IMS Presidential Address "Let us own data science", IMS Bulletin, 2014
Embracing statistical challenges in the IT age, Technometrics, 2007
In the news
IMS Wald Lecture I and Lecture II, and COPSS Distinguished Award and Lecture (DAAL) delieverd at Joint Statistical Meeting (JSM) in Toronto, August 2023
CDSS news: Statistics-Computer Science team reflects on tackling covid outbreaks, May, 2022
Honorary Doctorate, University of Lausanne (UNIL) (Faculty of Business and Economics), June 4, 2021 (Interview of Bin Yu by journalist Nathalie Randin, with an introduction by Dean Jean-Philippe Bonardi of UNIL in French (English translation))
CDSS news on our PCS framework: "A better framework for more robust, trustworthy data science", Oct. 2020
UC Berkeley to lead $10M NSF/Simons Foundation program to investigate theoretical underpinnings of deep learning, Aug. 25, 2020
Curating COVID-19 data repository and forecasting county-level death counts in the US, 2020
Seeking Data Wisdom, 2015
One of the 50 best inventions of 2011 by Time Magazine, 2011