Dr. Mimi Zhang

Mimi Zhang joined TCD as an assistant professor in October 2017. She holds a B.Sc. in statistics from University of Science and Technology of China (Sep. 2007-Jul. 2011), and a Ph.D. in industrial engineering from City University of Hong Kong (Nov. 2011-Dec. 2014). Before joining TCD, she was a research associate at University of Strathclyde and Imperial College London.
Her main research areas are machine learning and operations research, including cluster analysis, Bayesian optimization, functional data analysis, reliability & maintenance (engineering), etc. Her collaborations primarily span the fields of mechanical, manufacturing, and biomedical engineering. She is the strand leader of the Data Science MSc programme and an AE for Journal of Classification.

Current PhD students:
My research draws on advanced mathematics and statistical techniques. Therefore, I only consider PhD candidates with a strong background in mathematics, statistics, or computer science (not computer engineering).

• Guangchen Wang, 2023
• Samuel Singh, 2023
• Emmanuel Akeweje, 2023
• Jessica Bagnall, 2023, co-supervisor
• Sukriti Dhang, 2022, co-supervisor

Former PhD students:

• Joshua Tobin, thesis title "Consistent Mode-Finding for Parametric and Non-Parametric Clustering".
• Bernard Fares (part time), thesis title "Incorporating Ignorance within Game Theory: An Imprecise Probability Approach".

Teaching Activities

• 09/21-now: Introduction to Statistical Concepts and Methods (10 ECTS), Coordinator
• 09/21-now: Implementing Statistical Methods in R (5 ECTS), Coordinator
• 09/17-now: Software Application (5 ECTS), Coordinator
• 09/17-08/21: Statistics Base Module (15 ECTS), Coordinator

Software

• CPFcluster:   https://pypi.org/project/CPFcluster/
• REMclust:   https://pypi.org/project/REMclust/
• GPmix:   https://github.com/EAkeweje/GPmix

My academic journey spans from a foundation in mathematical statistics during my undergraduate studies to a focus on optimization algorithms and their applications in my doctoral and postdoctoral research. This interdisciplinary background integrates mathematics, probability, statistics, and algorithms to address diverse challenges across sectors like manufacturing, materials science, and healthcare.

Since becoming an independent researcher, my primary focus has been on cluster analysis, where I specialize in developing methodological, theoretical, and computational approaches for analyzing diverse data types, including multivariate, functional, and image data. In particular, functional data clustering aims to identify patterns across subjects, where each subject is represented by a continuous function. This technique has broad applications across various fields, such as grouping gene expression profiles in bioinformatics, economic time series in econometrics, and mechanical system vibrations in engineering.

Complementing my work in cluster analysis, my research portfolio extends to Bayesian Optimization -- a methodology designed to find the maximum (or minimum) of an unknown function, which is typically expensive to evaluate. The goal is to iteratively select the next best point to evaluate in order to efficiently search for the optimal solution. My collaborations in Bayesian optimization with academic and industry partners have afforded me the opportunity to address real-world challenges, a pursuit that I find immensely rewarding and fulfilling.