About the project
The project focuses on the theory, algorithms and applications for Machine Learning (ML) based on generalised nonconvex low-rank models. The goals include designing faster algorithm, understanding the working principle behind ML, applying ML to solve application problems.
We are seeking highly qualified candidates for doctoral positions on the research of ML.
This research is a mathematics-heavy project, it is not just about “applying method X on data Y”. You are expected to learn, understand, and develop theory to prove that “why method X will work”, “when will method X work” and “how fast will method X converge”.
We are looking for a MSc in applied mathematics, computer science, operations research, statistics, engineering, mathematics, or a related discipline with strong theoretical training.
You will have the flexibility to choose a topic within the range of this project.
- nonsmooth nonconvex optimisation
- optimisation on manifold
- analysis of optimisation algorithm by differential equations
- submodularity and combinatorial optimisation
- graph-theoretic machine learning
- numerical analysis / numerical linear algebra
- tensor-based signal processing
- optimal transport for machine learning
- foundation of machine learning via Clifford-Grassmann algebra
- foundation of machine learning via subtropical algebra
What we offer
- High quality training to do theoretical machine learning
- Exciting theoretical research topics
- Flexible research environment
Ideally you'll have:
- good mathematical skills
- good programming skills in a numerical language (such as MATLAB, Python, or Julia)
It would be good if you have:
- good communications skills, both written and oral, in English.
Before you make a decision, visit Dr Andersen Ang's personal web page for more information about the project.
