More Brain-like Computers Could Cut IT Energy Costs

Anyone who has witnessed the majestic and mesmerising flight of a murmuration of starlings has no doubt wondered how a flock of birds can achieve such synchronised behaviour. This is an example of emergence, where the interactions of simple things lead to complex collective behaviours. But emergence doesn't only occur in the natural world, and a group at the University of Sheffield is investigating how the emergent behaviour can be engineered in magnetic materials when they are patterned to have nanoscale dimensions.

Dr Tom Hayward, Senior Lecturer in Materials Physics at the University of Sheffield and author of this paper says,

Magnets have a number of properties that make them interesting for these kinds of applications: 

• Firstly, they are non-volatile, with inherent memory - if you stick a magnet to your fridge, it stays put.
• Brains (and brain-like computers) need to have non-linear responses, taking simple information and performing complicated transforms, and that's something magnets are naturally good at.
• There are plenty of ways to make magnets change state and perform computations that use very little energy.
• And magnets are a well-established technology (used, for example, in hard drives and Magnetoresistive random-access memory (MRAM)), and so there are existing routes to technology integration.

Key to this research is understanding what's happening to these magnetic nanorings when they're connected together - the way that emergence changes the way they change magnetic states.

Dr Hayward explains;

Schematic diagram showing three different reservoir architectures (a–c), with differing methods for providing input data (red circles) into reservoir nodes (blue circles) and reading reservoir state as output (green circles).

Communications Physics (Commun Phys) ISSN 2399-3650 (online) (CC BY)

The high-quality data the team collected at I06 allowed PhD candidate Ian Vidamour to gain a fundamental understanding of how the material behaved and develop ways to input and output data that "tune" the system to give very different computational properties from the same device.

After adding electrical contacts to the metamaterial to build an electrical device, the researchers used magnetic fields to input data and electrical resistances as the output, demonstrating state-of-the-art performance in diverse benchmarking computational tasks. 

It's all about connection

Developing a reconfigurable device like this raises the question of what happens when several of them - tuned to express different properties - are networked together. The brain has functional units contributing different things to its overall behaviour; will we see another level of emergence from these connected devices, so that they become more than the sum of their parts?

The team, which includes postdocs Charles Swindells and Guru Venkat, is also developing a single device with multiple inputs and outputs, and looking into more energy-efficient ways of driving the dynamic behaviour that don't rely on large magnets.

Another exciting area of research is the potential for using these magnetic materials as smart sensors.

Dr Hayward adds;

To find out more about the I06 beamline or discuss potential applications, please contact Principal Beamline Scientist Larissa Ishibe-Veiga: [email protected].

References:

¹ nnlabs.org. Power Requirements to Train Modern LLMs (2023).