Professor Albert Zomaya

Engineering and Information Technologies

Professor Liaquat Hossain

Engineering and Information Technologies

Professor Salah Sukkarieh

Engineering and Information Technologies

Self-assembly is the fundamental process that creates the specific conditions under which proteins, ecosystems, and societies spontaneously arrange themselves into a final entity (immune system, soil, consumers, etc.).

On the other hand, self-organisation is a process in which the components (or state) of a system increase in complexity without external intervention. In some cases, self-organising systems could exhibit emergent behaviour.

The aim of this project is use a range of algorithmic techniques (graph theoretic, game theoretic, information theoretic) to study and gain insight into the processes of self-assembly and self-organisation for a wide range of systems.

Moreover, the project will attempt to explore the complex nature of computation in systems that consist of hundreds or thousands of components (e.g. networks in biology, medicine, ecosystems, trade networks, etc.).

Register your interest

Optimisation algorithms can be used to solve a wide range of problems that arise in the environment-food-health system (e.g., policy and economic intervention strategies, environmental regeneration through use).

However, the many classical optimisation techniques (e.g., linear programming) are not suited for solving parallel processing problems due to their restricted nature. This project is investigating the application of unorthodox optimisation techniques such fuzzy logic, genetic algorithms, neural networks, simulated annealing, ant colonies, Tabu search, and others.

However, these techniques are computationally intensive and require enormous computing time. Parallel processing architectures (GPUs, FPGAs, Multicores, Clusters, etc.) have the potential for reducing the computational load and enabling the efficient use of these techniques to solve a wide variety of problems.

Register your interest

Transcription factors and miRNAs target different levels of regulation (transcription vs. translation) models that examine the reasons why these different control systems evolved and what utility they might have individually and collectively in maintaining health, would provide a useful framework for further biological enquiry.

The metaphor 'leader of the orchestra' has often been used in biological systems to describe master regulators of complex cellular events or of development. Transcription factors, which work alone or in combinatorial systems to promote or prevent gene expression in systems linking nutrition to health, have often been thought of as key 'conductors'.

More recently a new class of regulation has emerged by non-coding RNA molecules. There is only a limited number of these molecules (of the order of hundreds) identified, but each has targets in a number of genes, primarily regulating their translation.

Register your interest

As distributed systems that underpin a healthy lifestyle (e.g. social media, food distribution, socio-economic systems) become more prevalent with time there will be a need to endow such systems with capabilities that enable them to function optimally and adapt to new knowledge, social change and variable user needs.

 What makes this problem very complex is the heterogeneous nature of distributed environments (grids, clouds, trade networks, etc.) that could be made up of hundreds or thousands of components (computers, traders, citizens etc.).

In addition, no group in one location has control over other parts of the system. So there is a need to build resilience into these systems that allows them to adapt and maintain relevance.

Register your interest

For continued availability of food and water, much future science in environment modelling will require access to real-time measurements of a variety of observables across large spatial and temporal domains.

A number of large sensor networks are being developed to provide such measurements; of note are the NCRIS programs such as the Integrated Marine Observation System (IMOS) and the Terrestrial Ecosystem Research Network (TERN), as well as international programs such as the NSF Centre for Embedded Network Systems (CENS) Terrestrial Ecology Observing System (TEOS).

A key challenge in these large-scale networks is to assimilate and merge the information obtained to allow estimation and inference about parameters and models of interest. This project will aim to explore the use of distributed Bayesian networks as underlying models for these large distributed measurement systems which provide a direct method of estimation and inference.

The project will draw together science experts to provide domain knowledge for the inference process, statistical modellers and computer scientists to develop algorithms, such as those based on factor graphs and junction-tree techniques, for efficient computational approaches to these large-scale inference problems.

The project will propose a small number of domains to develop these ideas; possibly on the basis of the information being produced by the large NCRIS projects. The outcome of the project will be in the form of large-scale real-time inference techniques and potentially a set of tools for building inference engines for large critical sensor networks.

Register your interest

Many natural networks show remarkable resilience to failure of individual components, either through decentralisation of function or through adaptation and repair. What are the control strategies for such systems and how is information shared amongst components to provide this degree of robustness and integrity?

We propose to develop a simplified network model for such systems and to explore these issues of control and information sharing and their effects on overall system robustness. One hypothesis is that subsets of components contain all network information, used both to replicate lost information and to enable repair of the local network structure.

A second hypothesis is that the control structure allows different components to be co-opted and adapted to perform multiple purposes. The project would draw on two or three network systems; one based on a biological system, one or more on networks associated with public health.

A key outcome is to elucidate architectures that convey integrity and to determine how systems might be improved or how interventions should be designed in a more optimal way.

Register your interest

Network effects on individual's behaviour have been documented in studies on communications, sociology and social psychology. Previous studies demonstrate that actors with a dense social networks can behave more optimally.

Furthermore, actors who are rich in structural holes (connections to social clusters or groups who are themselves not well connected) are better situated in their social network to obtain, control and broker information. Task-oriented and sociological effects of information and communication technology use continue to be important in studies on individual behaviour, including the choices individuals make in relation to their health.

Research on direct interplay between social network structure, information and individual behaviour is, however, lacking to date. This project will study a wide range of biological and social networks with a view to understanding how network structure can be modified to improve desired outcomes.

Register your interest