Bin Yu

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).

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.

Veridical data science (PCS framework), PNAS, 2020 (QnAs with Bin Yu)

Breiman Lecture (video) at NeurIPS "Veridical data Science" (PCS framework and iRF), 2019; updated slides, 2020

Definitions, methods and applications in interpretable machine learning, PNAS, 2019

Data wisdom for data science (blog), 2015

IMS Presidential Address "Let us own data science", IMS Bulletin, 2014

Stability, Bernoulli, 2013

Embracing statistical challenges in the IT age, Technometrics, 2007

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

Interviewed by PBS Nova about AlphaZero, 2018

Mapping a cell's destiny, 2016

Seeking Data Wisdom, 2015

Member, National Academy of Sciences, 2014

Fellow, American Academy of Arts and Sciences, 2013

One of the 50 best inventions of 2011 by Time Magazine, 2011

The Economist Article, 2011

ScienceMatters @ Berkeley. Dealing with Cloudy Data, 2004

See original here:
Bin Yu

Related Posts

Comments are closed.