Jiawei Zhang | University of Kentucky College of Arts & Sciences

Jiawei Zhang

Contact Information

jiawei.zhang@uky.edu

MDS 337

(859) 257-6912

Research Interests

Machine learning, privacy-aware decentralized learning, deep learning, statistical inference, model selection and diagnostics

Affiliations

Statistics

Website

______________________________________________________________________

I have been working on model diagnostics, model improvement, and model selection, which are three important and interconnected aspects of machine learning. For model diagnostics, I have developed a goodness-of-fit assessment tool to check whether a classifier has systematic defects. For model improvement, I have proposed a learning framework to help data analysts improve their models by leveraging external resources, including external computation power and datasets. For model selection, I have introduced a method that emphasizes areas of particular interest, for instance, it may select the best disease diagnostic model for some minority groups.

Selected Publications

Selected Publications:

J. Zhang, J. Ding, Y. Yang, “Is a Classification Procedure Good Enough? A Goodness‐of‐Fit Assessment Tool for Classification Learning,” Journal of the American Statistical Association, (2023)
J.Zhang,J.Ding,Y.Yang,“TargetCross‐Validation,”Bernoulli,(2023)
X. Wang, J. Zhang, M. Hong, Y. Yang, J. Ding, “Parallel Assisted Learning,” IEEE Transactions on Signal Processing, (2022)
J. Zhang, Y. Yang, J. Ding, “Information Criteria for Model Selection,” Wiley Interdisciplinary Reviews: Computational Statistics, (2023)
X.Tang,J.Zhang(co‐firstauthor),Y.He,X.Zhang,Z.Lin,S.Partarrieu,E.B.Hanna,Z.Ren,Y.Yang,X.Wang,N.

Li, J. Ding, J. Liu, “Multi‐task learning for single‐cell multi‐modality biology,” Nature Communications, (2023)
E. Diao, G. Wang, J. Zhang, Y. Yang, J. Ding, V. Tarokh, “Pruning Deep Neural Networks from a Sparsity Perspective,” ICLR 2023, (2023)
C. Chen, J. Zhang (co‐first author), J. Ding, Y. Zhou, “Assisted Learning with Unsupervised Domain Adaptation,” Abstracts of papers IEEE International Symposium on Information Theory 2, (2023)