arXiv subjects

Matthew O A Ellis

Publications and source records attributed to Matthew O A Ellis.

5 recordsLinked to original sources

arXiv2026

Lightweight Test-Time Adaptation for EMG-Based Gesture Recognition

Reliable long-term decoding of gestures from surface electromyography (EMG) is hindered by signal drift caused by electrode displacement, muscle fatigue, and/or posture changes. Although modern models achieve high intra-session accuracy, their performance often degrades substantially across recording sessions. Existing approaches to mitigate this problem typically rely on large training datasets or computationally intensive pipelines that are unsuitable for energy-efficient wearable devices. We propose a lightweight test-time adaptation framework for EMG decoding. The framework includes three complementary adaptation strategies: (i) causal adaptive batch normalization for online statistical alignment, (ii) Gaussian Mixture Model alignment with experience replay to mitigate forgetting, and (iii) meta-learning for rapid few-shot calibration. We evaluate these methods on the multi-session NinaPro DB6 dataset. All approaches substantially improve inter-session robustness relative to a non-adaptive baseline while maintaining low computational overhead. Replay-regularized statistical alignment provides the most stable adaptation under limited data, while meta-learning achieves the highest accuracy when sparse calibration labels are available. Overall, our self-supervised test-time adaptation methods reach up to 82% inter-session accuracy, significantly improving upon prior approaches while maintaining resource-efficient operation. These results demonstrate that lightweight test-time adaptation can enable robust, long-term EMG decoding for wearable or prosthetic applications.

cs.LG↗

arXiv2023

A route to minimally dissipative switching in magnets via THz phonon pumping

Advanced magnetic recording paradigms typically use large temperature changes to drive switching which is detrimental to device longevity, hence finding non-thermal routes is crucial for future applications. By employing atomistic spin-lattice dynamics simulations, we show efficient coherent magnetisation switching triggered by THz phonon excitation in insulating single species materials. The key ingredient is excitation near the $P$-point of the spectrum in conditions where spins typically cannot be excited and when manifold $k$ phonon modes are accessible at the same frequency. Our model predicts the necessary ingredients for low-dissipative switching and provides new insight into THz-excited spin dynamics.

cond-mat.mtrl-sci↗

arXiv2022

A perspective on physical reservoir computing with nanomagnetic devices

Neural networks have revolutionized the area of artificial intelligence and introduced transformative applications to almost every scientific field and industry. However, this success comes at a great price; the energy requirements for training advanced models are unsustainable. One promising way to address this pressing issue is by developing low-energy neuromorphic hardware that directly supports the algorithm's requirements. The intrinsic non-volatility, non-linearity, and memory of spintronic devices make them appealing candidates for neuromorphic devices. Here we focus on the reservoir computing paradigm, a recurrent network with a simple training algorithm suitable for computation with spintronic devices since they can provide the properties of non-linearity and memory. We review technologies and methods for developing neuromorphic spintronic devices and conclude with critical open issues to address before such devices become widely used.

cs.ET↗

arXiv2022

Quantifying the Computational Capability of a Nanomagnetic Reservoir Computing Platform with Emergent Magnetization Dynamics

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.

cond-mat.mes-hall↗

arXiv2020

Modeling coupled spin and lattice dynamics

A unified model of molecular and atomistic spin dynamics is presented enabling simulations both in microcanonical and canonical ensembles without the necessity of additional phenomenological spin damping. Transfer of energy and angular momentum between the lattice and the spin systems is achieved by a coupling term based upon the spin-orbit interaction. The characteristic spectra of the spin and phonon systems are analyzed for different coupling strength and temperatures. The spin spectral density shows magnon modes together with the uncorrelated noise induced by the coupling to the lattice. The effective damping parameter is investigated showing an increase with both coupling strength and temperature. The model paves the way to understanding magnetic relaxation processes beyond the phenomenological approach of the Gilbert damping and the dynamics of the energy transfer between lattice and spins.

cond-mat.mtrl-sci↗