Data science is about the automatic extraction of knowledge from data, using methods from
Statistics, Maths, and Computer Science. Applications are wide, e..g, from smart electrical energy, environmental monitoring, social networks, to intelligent materials and robots, cyber-security or preventative health.
My research interest is in machine learning and intelligent & complex systems, with a focus on recurrent neural networks for prediction and classification. Recurrent neural networks are particularly well suited for sequential and time-series data. I am investigating efficient training methods, and also ways to distribute computation in large neural networks.
Artificial neural network approaches take inspiration from information processing in brains, and mimicking self-organised mechanisms for learning may help to develop successful learning approaches that can be easily distributed, or implemented on non-conventional hardware (e.g., using chemistry). To investigate and compare computational properties of this self-organisation, we use and extend approaches from complex systems research and information theory (information dynamics), with the goal to guide this self-organisation into a desired direction. I am also interested in effects that artificial neural networks may inherit from their biological counterparts, like visual illusions.
Publications, Projects, Code, News and more information
The links in the menu above lead to more information about my publications, selected projects, academic activities, and about some code that I have produced. I also try to maintain a blog about some of these activities (Evil Robots, an anagram of my name :).
Supervision and student projects
Current or prospective students at Western Sydney University are welcome to contact me for research / PhD projects. If you are looking for a job or a paid internship, please check the University page for any open positions – I usually do not reply to unsolicited applications.