A 2025 panel discussion, including Bin Yu, a researcher at the Center for Information Technology Research in the Interest of Society and the Banatao Institute (CITRIS) and professor in the Departments of Statistics and of Electrical Engineering and Computer Sciences at the University of California Berkeley, was published April 8 in the Journal of Statistics and Data Science Education.
During the panel, researchers discussed veridical data science, an approach that emphasizes real-world validity, using representative data, rigorous and transparent methods, and evaluation practices that reflect how systems perform in practice. The discussion covered how the field is being adopted, panelists’ paths to data science, the evolving goals of data science education, ongoing technical and institutional challenges and more.
“The key goal is to teach a way of doing data science, which is holistic, through the current methodologies and practical projects in the context of domain knowledge,” Yu said during the panel.
Yu co-authored a 2024 book on veridical data science and received COVID 2020 CITRIS seed funding for a project which aimed to develop COVID-19 forecasting methods. Currently, she researches statistics and machine learning theory, methodologies, and algorithms for solving high-dimensional data problems.
Read more in the Journal of Statistics and Data Science Education.