Magnetic Architectures for Reservoir Computing Hardware (MARCH)

Unconventional computers such as in materio `Reservoir Computers' (RCs) can help overcome these issues, particularly by being able to perform computation that directly exploits the natural dynamics of their materials.  Such devices have been proven in principle, but systems need to be scaled up to provide sufficient and appropriate computational power for real world tasks. A diverse range of device configurations are needed to support different computational tasks.

The MARCH project's cross-disciplinary team from York (CS/NSC and EE) and Sheffield (CS and Materials Science), funded by UKRI/EPSRC, is collaborating to develop novel systems comprising multiple coupled magnetic material RCs.

Magnetic materials provide an excellent flexible testbed for developing a more general material RC design process. Such materials have the intrinsic memory and complex non-linear dynamics needed for RC operation, and also have well-established methods of interfacing for data input/output, as needed to build a practical device. We are exploiting the properties of patterned 2D layouts of magnetic nanoring wires, which we can manufacture with existing technologies. This flexible design and experimental platform allows us to develop generic techniques that will apply across a range of smart material RCs.

The output will include a new design methodology and platform for multi-reservoir devices, that can be exploited to design low-power, robust, flexible, and efficient smart sensing and other `edge-computing' devices in a diverse range of materials.