Imagine being able to predict when the next big earthquake or volcanic eruption will happen so you can warn people in advance. Or determining an individual’s risk of developing Alzheimer’s disease in the future. Or knowing which choice a person will make before they know themselves.
These are just a few of the artificial intelligence (AI) projects being undertaken by researchers at AUT who are part of the University’s wide-reaching AI initiative.
Underpinning a number of these AI projects is a groundbreaking machine learning system called NeuCube, developed by AUT Professor Nik Kasabov, that mimics the way our brains learn and recognise patterns.
Massive investment is being made in AI around the world but, even with intense international competition, the NeuCube and Nik and his team’s research are capturing global attention.
“At AUT’s Knowledge Engineering and Discovery Research Institute (KEDRI) we are focused on new AI methods and technologies called ‘Brain-Inspired AI’,” says Nik, who is the director of KEDRI and an advisory professor at Shanghai Jiao Tong University in China.
“NeuCube allows for brain data to be mapped, learned and understood, along with other types of spatiotemporal data – for example seismic activity or an individual’s health and environmental factors at a given point in time.”
NeuCube is already being used in 25 labs around the world and KEDRI researchers are working on a number of major international AI projects with the EU and China. This includes the $2.2m research exchange project, Pacific Atlantic Network for Technical Higher Education and Research (PANTHER) which aims to develop research collaboration between the European Union, Australian and New Zealand universities with a focus on engineering and technology.
In AUT’s own labs, postgraduate students like PhD candidates Zohreh and Maryam Doborjeh are making breakthroughs of their own, under the guidance of their supervisors.
In one experiment, participants were asked to watch a video of different beverage logos while their brain data was recorded using an EEG headset, the data then synced with the computational power of the NeuCube.
“NeuCube allows for brain data to be mapped, learned and understood, along with other types of spatio-temporal data – for example seismic activity or an individual’s health and environmental factors at a given point in time.”
The result? It was possible to predict a participant’s subconscious decision, such as beverage choice, 0.2 seconds after the presentation of the stimuli, which is before they even consciously perceived the beverage.
Maryam, who specialises in machine learning, says witnessing the NeuCube algorithm work was amazing.
“The brain is an amazing thing – it learns and remembers things and can recognise them before the person can. To get a computer to be able to do that will change the way we all live.”