Job Description

The successful candidate will work with Dr. Doudou Zhou on the development of novel statistical and machine learning methods for analyzing complex modern data under a project on "Novel Methods for Analyzing High-Dimensional and Non-Euclidean Data".

This project aims to develop rigorous, scalable, and interpretable methods for real-world healthcare datasets such as electronic health records (EHR), which are often high-dimensional, multi-source, multi-modal, and incomplete. Methodological focus areas include reinforcement learning, transfer learning, multi-modal learning, high-dimensional statistics, graph neural network, change-point detection.

The main responsibilities of the position include:
1.    Conducting original research on statistical inference and machine learning for high-dimensional and/or multi-source data;
2.    Developing and implementing algorithms for federated learning, generative modelling, and representation learning;
3.    Preparing manuscripts for top-tier journals and conferences in statistics, machine learning, or biomedical informatics;
4.    Contributing to mentoring graduate or undergraduate students when appropriate;
5.    Assisting with grant reporting and collaborative project coordination.

Qualifications / Discipline:

•    A Ph.D. in Statistics, Biostatistics, Computer Science, or a related quantitative discipline;
•    Strong foundation in statistical theory, machine learning, or computational methods.
•    Experience working with real-world biomedical or healthcare data is an advantage.
•    A demonstrated record of academic publications.

Skills:
•    Proficiency in programming languages such as Python or R.
•    Ability to design and implement statistical or machine learning algorithms.
•    Excellent analytical and problem-solving skills.
•    Strong written and verbal communication skills.
•    Ability to work independently and collaboratively in a multidisciplinary environment.

Experience:
•    Prior research experience in at least one of the following areas would be preferred: high-dimensional statistics, reinforcement learning, multi-source/modal data, electronic health record (EHR) data analysis, or federated learning.
•    A strong publication record (or demonstrated potential) in peer-reviewed journals or top-tier conferences.
•    Experience working with real-world biomedical or healthcare data is an advantage.

More Information

Location: Kent Ridge Campus

Organization: Science

Department : Statistics and Data Science

Job requisition ID : 31566