Devices based on arrays of interconnected magnetic nano-rings with emergent magnetization dynamics have recently been proposed for use in reservoir computing applications, but for them to be computationally useful it must be possible to optimise their dynamical responses. Here, we use a phenomenological model to demonstrate that such reservoirs can be optimised for classification tasks by tuning hyperparameters that control the scaling and input-rate of data into the system using rotating magnetic fields. We use task-independent metrics to assess the rings’ computational capabilities at each set of these hyperparameters and show how these metrics correlate directly to performance in spoken and written digit recognition tasks. We then show that these metrics, and performance in tasks, can be further improved by expanding the reservoir’s output to include multiple, concurrent measures of the ring arrays’ magnetic states.
doi:10.1088/1361-6528/ac87b5 (open access)
@article(Vidamour++:2022, author = "Ian T. Vidamour and Matthew O. A. Ellis and David Griffin and Guru Venkat and Charles Swindells and Richard W. S. Dawidek and Thomas J. Broomhall and Nina-Juliane Steinke and Joshaniel F. K. Cooper and Francisco Maccherozzi and Sarnjeet S. Dhesi and Susan Stepney and Eleni Vasilaki and Dan A. Allwood and Thomas J. Hayward", title = "Quantifying the computational capability of a nanomagnetic reservoir computing platform with emergent magnetisation dynamics", volume = "33", pages = "485203", doi = "10.1088/1361-6528/ac87b5", journal = "Nanotechnology" )