Guosheng Yin (; born in 1976) is a statistician, data scientist, educator and researcher in Biostatistics, Statistics, machine learning, and AI. Presently, Guosheng Yin is Chair in Statistics in Department of Mathematics at Imperial College London. Previously, he served as the Head of Department and the Patrick S C Poon Endowed Chair in Statistics and Actuarial Science, at the University of Hong Kong. Before he joined the University of Hong Kong, Yin worked at the University of Texas M.D. Anderson Cancer Center till 2009 as a tenured Associate Professor of Biostatistics.

As an active researcher, he has published two books and over 220 scientific manuscripts in various peer-reviewed journals and conferences.

Guosheng Yin

Education

Guosheng Yin obtained his Bachelor's degree (BS) in Physics at Jilin University (Changchun, China) in 1995, Master's Degree (MA) in Physics at Temple University (PA, USA) in 1997 and earned Ph.D. in Biostatistics from the University of North Carolina at Chapel Hill (NC, USA) in 2003.

Guosheng Yin

Research and career

Yin's research interests include AI, deep learning, machine learning, Bayesian methods, clinical trial methodologies, adaptive designs, survival analysis, high-dimensional data and change-point analysis. He started as an Assistant Professor and worked as a tenured Associate Professor of Biostatistics in the University of Texas M.D. Anderson Cancer Center. Then, he joined the Department of Statistics and Actuarial Science at the University of Hong Kong as an Associate Professor in 2009 and promoted to a tenured Professor in 2014. In 2017-2023, he served as the Head of Department of Statistics and Actuarial Science, and he became an endowed chair as the Patrick S C Poon Professor in Statistics and Actuarial Science in 2018.

Guosheng Yin

Yin pioneered new statistical techniques to refine precision medicine, design adaptive clinical trials, and facilitate drug development. His versatile designs for clinical trials include adaptive phase I dose-finding methods, drug-combination studies, designs with late-onset outcomes and phase I/II trial designs. His ground-breaking research has ushered in advanced regression models and computational algorithms, enabling statistical modeling and theory application. He developed Bayesian generalized method of moments, established the connection between Bayesian posterior probability and p-value in hypothesis testing and further made a reconcilable interpretation on the well-known Lindley paradox. Yin has included machine learning and artificial intelligence in his research to amplify the precision and applicability of statistical science in solving real-world problems. Recently, he led a research team that integrated radiography and computer vision to develop a digital online diagnostic system for COVID-19 based on chest CT scans.

Awards

He has won recognition from various premier institutions:

Fellow of Institute of Mathematical Statistics (IMS) (2021) World's top 1% of scientists by Clarivate Analytics (2015, 2019, 2021) Fellow of American Statistical Association (ASA) (2013) Elected Member of International Statistical Institute (ISI) (2012) James E. Grizzle Distinguished Alumni Award, Department of Biostatistics, University of North Carolina at Chapel Hill (2009)

Publications

Yin is the author of two books and numerous research papers with more than 220 peer-reviewed publications in international scientific journals and conferences; the following list names a few of them:

Statistical Methodologies

PCA rerandomization. Canadian Journal of Statistics. 2023

Bayesian log-rank test. The American Statistician. 2023. CFO: Calibration-free odds design for phase I/II clinical trials. Statistical Methods in Medical Research 2022. Crystallization learning with the Delaunay triangulation. International Conference on Machine Learning ICML 2021. Demystify Lindley's paradox by connecting p-value and posterior probability. Statistics and Its Interface 2021. Reconnecting p-value and posterior probability under one- and two-sided tests. The American Statistician 2021, DOI: 10.1080/00031305.2020.1717621 Online COVID-19 diagnosis with chest CT images: Lesion-attention deep neural networks. 26th ACM SIGKDD Conference, 2020. Bayesian model averaging continual reassessment method in phase I clinical trials. Journal of American Statistical Association 104, 954–968, 2009. Bayesian generalized method of moments (with discussion). Bayesian Analysis 4, 191–222, 2009.