Adaptive Programmable Networks for In Materia Neuromorphic Computing

Stenning, Kilian D.; Gartside, Jack C.; Manneschi, Luca; Cheung, Christopher T. S.; Chen, Tony; Vanstone, Alex; Love, Jake; Holder, Holly H.; Caravelli, Francesco; Everschor-Sitte, Karin; Vasilaki, Eleni; Branford, W. R.

doi:10.48550/arxiv.2211.06373

Cited by 6 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In reservoir computing, a fixed reservoir performs a non-linear spatial-temporal transformation of an input sequence such that the output representation is linearly separable. The advantage of RC is that the reservoir transform can be offloaded to a physical system with appropriate properties and there has been considerable recent interest in developing magnetic (spintronics) based physical reservoir computing [7,[59][60][61][62][63][64][65][66]. There is potential to connect our magnetic DW based neural network to these reservoirs to create a complete hardware reservoir computing system.…”

Section: Discussionmentioning

confidence: 99%

Machine learning using magnetic stochastic synapses

Ellis

Welbourne

Kyle

et al. 2023

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

The impressive performance of artificial neural networks has come at the cost of high energy usage and CO2 emissions. Unconventional computing architectures, with magnetic systems as a candidate, have potential as alternative energy-efficient hardware, but, still face challenges, such as stochastic behaviour, in implementation. Here, we present a methodology for exploiting the traditionally detrimental stochastic effects in magnetic domain-wall motion in nanowires. We demonstrate functional binary stochastic synapses alongside a gradient learning rule that allows their training with applicability to a range of stochastic systems. The rule, utilising the mean and variance of the neuronal output distribution, finds a trade-off between synaptic stochasticity and energy efficiency depending on the number of measurements of each synapse. For single measurements, the rule results in binary synapses with minimal stochasticity, sacrificing potential performance for robustness. For multiple measurements, synaptic distributions are broad, approximating better-performing continuous synapses. This observation allows us to choose design principles depending on the desired performance and the device's operational speed and energy cost. We verify performance on physical hardware, showing it is comparable to a standard neural network.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine learning using magnetic stochastic synapses

Ellis

Welbourne

Kyle

et al. 2023

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

show abstract

“…High quality growth of films for low damping, may show best performance at low temperatures Artificial spin ices Time series prediction [200] (Type:Expt), multi-level and deep RC networks [218] (Type:Expt), Mackey-Glass and sunspot processing prediction [219] (Type:Expt) and pattern classification [207] (Type:Sim)…”

Section: Spin Torque Nano-oscillators and Superparamagnetic Tunnel Ju...mentioning

confidence: 99%

Magnetic domain walls: types, processes and applications

Venkat,

Allwood,

Hayward

2023

J. Phys. D: Appl. Phys.

View full text Add to dashboard Cite

Domain walls (DWs) in magnetic nanowires are promising candidates for a variety of applications including Boolean/unconventional logic, memories, in-memory computing as well as magnetic sensors and biomagnetic implementations. They show rich physical behaviour and are controllable using a number of methods including magnetic fields, charge and spin currents and spin-orbit torques. In this review, we detail types of domain walls in ferromagnetic nanowires and describe processes of manipulating their state. We look at the state of the art of DW applications and give our take on the their current status, technological feasibility and challenges.

show abstract

“…Artificial spin systems comprising networks of strongly interacting nanomagnets serve as promising hosts for future information-processing technologies, including nanomagnetic logic, 27,28 neuromorphic computation, [29][30][31][32][33][34][35] and reconfigurable magnonics. [36][37][38][39][40][41][42][43] Information can be stored in the magnetization of a single nanomagnet or the magnetic configuration of the entire network (microstate), where collective microstate-dependent dynamics 36,40,44 may be harnessed to process information.…”

Section: Introductionmentioning

confidence: 99%

“…[36][37][38][39][40][41][42][43] Information can be stored in the magnetization of a single nanomagnet or the magnetic configuration of the entire network (microstate), where collective microstate-dependent dynamics 36,40,44 may be harnessed to process information. [31][32][33][34][35] Local nanomagnet switching has been achieved through diffraction-limited heat-assisted reversal, relying on global fields in conjunction with laser illumination 2,45 and by cumbersome field-assisted 46 and field-free scanning-probe 41,47,48 techniques. In the latter, the magnetic field from a magnetic tip breaks the nanomagnet symmetry, causing transient domain wall formation and asymmetric propagation, leading to switching.…”

Section: Introductionmentioning

confidence: 99%