Alex Vanstone
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alex Vanstone has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Jack C. Gartside
Imperial College London
4 shared publications - Kilian D. Stenning
Imperial College London
4 shared publications - Will R. Branford
Imperial College London
4 shared publications - Hidekazu Kurebayashi
London Centre for Nanotechnology, University College London, London, UK
3 shared publications - Troy Dion
Kyushu University
3 shared publications - Daan M. Arroo
Department of Materials, Imperial College London, London, UK
2 shared publications
Latest Publications
- Open Access
- 27 August 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-15
Neuromorphic overparameterisation and few-shot learning in multilayer physical neural networks
Physical neuromorphic computing, exploiting the complex dynamics of physical systems, has seen rapid advancements in sophistication and performance. Physical reservoir computing, a subset of neuromorphic computing, faces limitations due to its reliance on single systems. This constrains output dimensionality and dynamic range, limiting performance to a narrow range of tasks. Here, we engineer a suite of nanomagnetic array physical reservoirs and interconnect them in parallel and series to create a multilayer neural network architecture. The output of one reservoir is recorded, scaled and virtually fed as input to the next reservoir. This networked approach increases output dimensionality, internal dynamics and computational performance. We demonstrate that a physical neuromorphic system can achieve an overparameterised state, facilitating meta-learning on small training sets and yielding strong performance across a wide range of tasks. Our approach’s efficacy is further demonstrated through few-shot learning, where the system rapidly adapts to new tasks.
- 27 August 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-15
Open AccessNeuromorphic overparameterisation and few-shot learning in multilayer physical neural networks
Physical neuromorphic computing, exploiting the complex dynamics of physical systems, has seen rapid advancements in sophistication and performance. Physical reservoir computing, a subset of neuromorphic computing, faces limitations due to its reliance on single systems. This constrains output dimensionality and dynamic range, limiting performance to a narrow range of tasks. Here, we engineer a suite of nanomagnetic array physical reservoirs and interconnect them in parallel and series to create a multilayer neural network architecture. The output of one reservoir is recorded, scaled and virtually fed as input to the next reservoir. This networked approach increases output dimensionality, internal dynamics and computational performance. We demonstrate that a physical neuromorphic system can achieve an overparameterised state, facilitating meta-learning on small training sets and yielding strong performance across a wide range of tasks. Our approach’s efficacy is further demonstrated through few-shot learning, where the system rapidly adapts to new tasks.
- Open Access
- 14 May 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-13
Ultrastrong magnon-magnon coupling and chiral spin-texture control in a dipolar 3D multilayered artificial spin-vortex ice
Strongly-interacting nanomagnetic arrays are ideal systems for exploring reconfigurable magnonics. They provide huge microstate spaces and integrated solutions for storage and neuromorphic computing alongside GHz functionality. These systems may be broadly assessed by their range of reliably accessible states and the strength of magnon coupling phenomena and nonlinearities. Increasingly, nanomagnetic systems are expanding into three-dimensional architectures. This has enhanced the range of available magnetic microstates and functional behaviours, but engineering control over 3D states and dynamics remains challenging. Here, we introduce a 3D magnonic metamaterial composed from multilayered artificial spin ice nanoarrays. Comprising two magnetic layers separated by a non-magnetic spacer, each nanoisland may assume four macrospin or vortex states per magnetic layer. This creates a system with a rich 16N microstate space and intense static and dynamic dipolar magnetic coupling. The system exhibits a broad range of emergent phenomena driven by the strong inter-layer dipolar interaction, including ultrastrong magnon-magnon coupling with normalised coupling rates of $$\frac{\Delta f}{\nu }=0.57$$ , GHz mode shifts in zero applied field and chirality-control of magnetic vortex microstates with corresponding magnonic spectra.- 14 May 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-13
Open AccessUltrastrong magnon-magnon coupling and chiral spin-texture control in a dipolar 3D multilayered artificial spin-vortex ice
Strongly-interacting nanomagnetic arrays are ideal systems for exploring reconfigurable magnonics. They provide huge microstate spaces and integrated solutions for storage and neuromorphic computing alongside GHz functionality. These systems may be broadly assessed by their range of reliably accessible states and the strength of magnon coupling phenomena and nonlinearities. Increasingly, nanomagnetic systems are expanding into three-dimensional architectures. This has enhanced the range of available magnetic microstates and functional behaviours, but engineering control over 3D states and dynamics remains challenging. Here, we introduce a 3D magnonic metamaterial composed from multilayered artificial spin ice nanoarrays. Comprising two magnetic layers separated by a non-magnetic spacer, each nanoisland may assume four macrospin or vortex states per magnetic layer. This creates a system with a rich 16N microstate space and intense static and dynamic dipolar magnetic coupling. The system exhibits a broad range of emergent phenomena driven by the strong inter-layer dipolar interaction, including ultrastrong magnon-magnon coupling with normalised coupling rates of $$\frac{\Delta f}{\nu }=0.57$$ , GHz mode shifts in zero applied field and chirality-control of magnetic vortex microstates with corresponding magnonic spectra. - Open Access
- 1 April 2022
- Published by IOP Publishing in New Journal of Physics
- Vol. 24 (4) , 043017
Spectral fingerprinting: microstate readout via remanence ferromagnetic resonance in artificial spin ice
Artificial spin ices (ASIs) are magnetic metamaterials comprising geometrically tiled strongly-interacting nanomagnets. There is significant interest in these systems spanning the fundamental physics of many-body systems to potential applications in neuromorphic computation, logic, and recently reconfigurable magnonics. Magnonics focused studies on ASI have to date have focused on the in-field GHz spin-wave response, convoluting effects from applied field, nanofabrication imperfections (‘quenched disorder’) and microstate-dependent dipolar field landscapes. Here, we investigate zero-field measurements of the spin-wave response and demonstrate its ability to provide a ‘spectral fingerprint’ of the system microstate. Removing applied field allows deconvolution of distinct contributions to reversal dynamics from the spin-wave spectra, directly measuring dipolar field strength and quenched disorder as well as net magnetisation. We demonstrate the efficacy and sensitivity of this approach by measuring ASI in three microstates with identical (zero) magnetisation, indistinguishable via magnetometry. The zero-field spin-wave response provides distinct spectral fingerprints of each state, allowing rapid, scaleable microstate readout. As artificial spin systems progress toward device implementation, zero-field functionality is crucial to minimize the power consumption associated with electromagnets. Several proposed hardware neuromorphic computation schemes hinge on leveraging dynamic measurement of ASI microstates to perform computation for which spectral fingerprinting provides a potential solution.
- 1 April 2022
- Published by IOP Publishing in New Journal of Physics
- Vol. 24 (4) , 043017
Open AccessSpectral fingerprinting: microstate readout via remanence ferromagnetic resonance in artificial spin ice
Artificial spin ices (ASIs) are magnetic metamaterials comprising geometrically tiled strongly-interacting nanomagnets. There is significant interest in these systems spanning the fundamental physics of many-body systems to potential applications in neuromorphic computation, logic, and recently reconfigurable magnonics. Magnonics focused studies on ASI have to date have focused on the in-field GHz spin-wave response, convoluting effects from applied field, nanofabrication imperfections (‘quenched disorder’) and microstate-dependent dipolar field landscapes. Here, we investigate zero-field measurements of the spin-wave response and demonstrate its ability to provide a ‘spectral fingerprint’ of the system microstate. Removing applied field allows deconvolution of distinct contributions to reversal dynamics from the spin-wave spectra, directly measuring dipolar field strength and quenched disorder as well as net magnetisation. We demonstrate the efficacy and sensitivity of this approach by measuring ASI in three microstates with identical (zero) magnetisation, indistinguishable via magnetometry. The zero-field spin-wave response provides distinct spectral fingerprints of each state, allowing rapid, scaleable microstate readout. As artificial spin systems progress toward device implementation, zero-field functionality is crucial to minimize the power consumption associated with electromagnets. Several proposed hardware neuromorphic computation schemes hinge on leveraging dynamic measurement of ASI microstates to perform computation for which spectral fingerprinting provides a potential solution.
- Open Access
- 11 February 2022
- Published by American Physical Society (APS) in Physical Review Research
- Vol. 4 (1) , 013107
Observation and control of collective spin-wave mode hybridization in chevron arrays and in square, staircase, and brickwork artificial spin ices
Dipolar magnon-magnon coupling has long been predicted in nanopatterned artificial spin systems. However, observation of such phenomena and related collective spin-wave signatures have until recently proved elusive or been limited to low-power edge modes which are difficult to measure experimentally. Here we describe the requisite conditions for dipolar mode-hybridization, how it may be controlled, why it was not observed earlier, and how strong coupling may occur between nanomagnet bulk modes. We experimentally investigate four nanopatterned artificial spin system geometries: chevron arrays, square, staircase, and brickwork artificial spin ices. We observe significant dynamic dipolar-coupling in all systems with relative coupling strengths and avoided-crossing gaps supported by micromagnetic-simulation results. We demonstrate reconfigurable mode-hybridization regimes in each system via microstate control, and in doing so elucidate the underlying dynamics governing dynamic dipolar-coupling with implications across reconfigurable magnonics. We demonstrate that confinement of the bulk modes via edge effects plays a critical role in dipolar hybridized modes, and treating each nanoisland as a coherently precessing macro-spin or a standing spin-wave is insufficient to capture experimentally observed coupling phenomena. Finally, we present a parameter-space search detailing how coupling strength may be tuned via nanofabrication dimensions and material properties.
- 11 February 2022
- Published by American Physical Society (APS) in Physical Review Research
- Vol. 4 (1) , 013107
Open AccessObservation and control of collective spin-wave mode hybridization in chevron arrays and in square, staircase, and brickwork artificial spin ices
Dipolar magnon-magnon coupling has long been predicted in nanopatterned artificial spin systems. However, observation of such phenomena and related collective spin-wave signatures have until recently proved elusive or been limited to low-power edge modes which are difficult to measure experimentally. Here we describe the requisite conditions for dipolar mode-hybridization, how it may be controlled, why it was not observed earlier, and how strong coupling may occur between nanomagnet bulk modes. We experimentally investigate four nanopatterned artificial spin system geometries: chevron arrays, square, staircase, and brickwork artificial spin ices. We observe significant dynamic dipolar-coupling in all systems with relative coupling strengths and avoided-crossing gaps supported by micromagnetic-simulation results. We demonstrate reconfigurable mode-hybridization regimes in each system via microstate control, and in doing so elucidate the underlying dynamics governing dynamic dipolar-coupling with implications across reconfigurable magnonics. We demonstrate that confinement of the bulk modes via edge effects plays a critical role in dipolar hybridized modes, and treating each nanoisland as a coherently precessing macro-spin or a standing spin-wave is insufficient to capture experimentally observed coupling phenomena. Finally, we present a parameter-space search detailing how coupling strength may be tuned via nanofabrication dimensions and material properties.