Algorithms for Data Science (MSc), 2020 - present (module lead). 

Artificial Intelligence II (BSc), 2020 - present.

Mathematical Foundation of Artificial Intelligence and Machine Learning (MSc), 2020/21.

Artificial Intelligence, (BSc, MSc), 2019/20.

Student Projects

My research interests lie in evolutionary computation and multi-objective optimisation, and their applications. Students may choose a project in the areas list below, or propose a project which relates to my research interests.

• Develop randomised algorithms (e.g., genetic algorithms, local search, novelty search) for combinatorial optimisation problems. This includes classic and emerging problems:

  • TSP (Travelling Salesman Problem); 

  • KP (Knapsack Problem);

  • JSSP (Job Shop Scheduling Problem);

  • VRP (Vehicle Routing Problem); 

  • Timetabling;

  • Bin Packing; 

  • Allocation; 

  • NK-Landscape

  • Quadratic Assignment Problem

  • ...

• Their multi-objective version. Research projects include:

  • To design effective algorithms for a specific multi-objective combinatorial problem.

  • To investigate what kind of search algorithms perform well for a specific multi-objective combinatorial problem.

  • To investigate how well evolutionary algorithms scales up in one type of combinatorial problems.

  • To investigate the search operators (e.g., crossover and mutation) in one type of combinatorial problems and how well they scales up.

• Data visualisation in multi-objective optimisation. When the number of objectives to be optimised is larger than 3, it is not straightforward to observe the results. Therefore, it is important to develop effective methods/tools to help visualise the results. Related projects include:

  • Reducing the dimensionality according to correlation between objectives.

  • Modifying visualisation tools (e.g. parallel coordinates) for clearer observation.

• Multi-criterion decision making. Multi-criteria decision making is a decision making process that involving multiple conflicting criteria/objectives, It is ubiquitous in our daily life (e.g. when purchasing a car, cost, comfort, safety, and fuel economy may be some of the main criteria we consider - it is unusual that the cheapest car is the most comfortable and the safest one). Related projects include:

  • How to select a few representative solutions for better decision making.

  • How to make the results clearer for observation and decision making in high-dimensional space.

  • Comparison between different decision making approaches and their relevance to multi-objective optimisation.

• Comparison between randomised search techniques and MIP (mixed integer programming) like Cplex and Gurobi on a variety of problem scenarios.

• Evolution in art. Use population-based search algorithms (e.g., genetic algorithms) to discover novel/unexpected patterns and designs, for example, to generate novel tree patterns and Chinese-style patterns.