Quantifying the Computational Capability of a Nanomagnetic Reservoir Computing Platform with Emergent Magnetization Dynamics
This work addresses improving computational efficiency for reservoir computing applications in AI hardware, but it is incremental as it builds on existing methods with specific optimizations.
The researchers tackled optimizing nanomagnetic reservoir computing for classification by tuning hyperparameters like scaling and input rate, showing that task-independent metrics correlate with performance in digit recognition tasks and can be improved by expanding output measures.
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.