🤖 AI Summary
This work proposes a reservoir computing architecture based on geometrically heterogeneous arrays of magnetic nanorings to harness the intrinsic nonlinearity and history-dependent dynamics of physical systems for efficient temporal computation with minimal training overhead. Input signals are encoded via a rotating magnetic field, and the responses of nanorings with varying widths are simultaneously read out through multichannel planar Hall effect measurements. Geometric heterogeneity is introduced for the first time as a new degree of freedom to tailor reservoir dynamics, and combined with principal component analysis for noise suppression, substantially enhancing representational capacity. In Mackey-Glass time-series prediction tasks, the multichannel cooperative output significantly reduces the normalized root mean square error, demonstrating improved computational performance and paving the way toward scalable magnetic metamaterial-based computing systems.
📝 Abstract
Physical reservoir computing utilizes the intrinsic nonlinear and history-dependent dynamics of physical systems to perform machine-learning tasks with minimal training overhead. Here, we introduce a nanomagnetic reservoir computer based on a heterogeneous array of interconnected magnetic nanorings, combined with multi-channel planar Hall effect readout. The device comprises subarrays of rings with systematically varied track widths ranging from 500 nm to 300 nm, enabling access to the heterogeneous dynamics of geometrically diverse magnetic systems within a single reservoir. By applying time-varying input signals as modulations of a driving rotating magnetic field, we evaluate the nanoring reservoir's performance on nonlinear signal transformation and Mackey-Glass time-series prediction tasks. We find that combining outputs from multiple width-dependent channels significantly reduces the normalized root-mean-square error compared to single-channel readout, with the optimal channel combinations depending on task requirements. These results demonstrate that geometric heterogeneity provides an additional, experimentally accessible degree of freedom and complementary computational features. Principal component analysis further reveals that a reduced subset of correlated features captures most of the computationally relevant information while suppressing noise contributions. These results demonstrate that controlled geometric heterogeneity enhances reservoir expressivity and suggest a route toward scalable magnetic computing architectures in which multi-output magnetic metamaterials serve as configurable dynamical building blocks for device networks.