arXiv subjects

J. Grollier

Publications and source records attributed to J. Grollier.

At least 19 recordsLinked to original sources

arXiv2026

Remotely programming the weights of a spintronic neural network by a radiofrequency broadcast signal

Selectively programming large number of non-volatile synaptic weights without compromising scalability is a key challenge for in-memory computing. Here, we demonstrate remote programming of synaptic weights in series-connected chains of 11 vortex-based magnetic tunnel junctions using broadcast radiofrequency signals applied through a shared strip line. The programming relies on frequency-selective reversal of the vortex-core polarity and therefore does not require individual access lines or selector devices. By reconfiguring the binary states of these chains, we reshape the weighted sums they perform on frequency-multiplexed RF inputs. Using a 22-synapse network composed of two such chains, we remotely reconfigure the same hardware to perform two distinct tasks: handwritten-digit classification and drone RF-signature identification. The digit-optimized configuration reaches 94.91 +/- 0.26% accuracy on handwritten digits but only 13.17 +/- 0.47% on drone RF signatures, whereas the drone-optimized configuration reaches 97.33 +/- 0.62% on drones but only 47.59 +/- 1.5% on digits. Broadcast RF programming thus provides a compact and scalable route to rapidly reconfigurable spintronic neuromorphic hardware.

cs.ET↗

arXiv2024

Reconfigurable classifier based on spin torque driven magnetization switching in electrically connected magnetic tunnel junctions

A promising branch of neuromorphic computing aims to perform cognitive operations in hardware leveraging the physics of efficient and well-established nano-devices. In this work, we present a reconfigurable classifier based on a network of electrically connected magnetic tunnel junctions that categorizes information encoded in the amplitude of input currents through the spin torque driven magnetization switching output configuration. The network can be trained to classify new data by adjusting additional programming currents applied selectively to the junctions. We experimentally demonstrate that a network composed of three magnetic tunnel junctions can learn to classify spoken vowels with a recognition rate that surpasses the performance of software multilayer neural networks with the same number of trained parameters in this task. These results, obtained with the same nano-devices and working principle employed in industrial spin-transfer torque magnetic random-access memories (STT-MRAM), constitute an important step towards the development of large-scale neuromorphic networks based on well-established technology.

physics.app-ph↗

arXiv2021

Radio-Frequency Multiply-And-Accumulate Operations with Spintronic Synapses

Exploiting the physics of nanoelectronic devices is a major lead for implementing compact, fast, and energy efficient artificial intelligence. In this work, we propose an original road in this direction, where assemblies of spintronic resonators used as artificial synapses can classify an-alogue radio-frequency signals directly without digitalization. The resonators convert the ra-dio-frequency input signals into direct voltages through the spin-diode effect. In the process, they multiply the input signals by a synaptic weight, which depends on their resonance fre-quency. We demonstrate through physical simulations with parameters extracted from exper-imental devices that frequency-multiplexed assemblies of resonators implement the corner-stone operation of artificial neural networks, the Multiply-And-Accumulate (MAC), directly on microwave inputs. The results show that even with a non-ideal realistic model, the outputs obtained with our architecture remain comparable to that of a traditional MAC operation. Us-ing a conventional machine learning framework augmented with equations describing the physics of spintronic resonators, we train a single layer neural network to classify radio-fre-quency signals encoding 8x8 pixel handwritten digits pictures. The spintronic neural network recognizes the digits with an accuracy of 99.96 %, equivalent to purely software neural net-works. This MAC implementation offers a promising solution for fast, low-power radio-fre-quency classification applications, and a new building block for spintronic deep neural net-works.

cond-mat.dis-nn↗

arXiv2020

Spintronics for neuromorphic computing

Neuromorphic computing uses brain-inspired principles to design circuits that can perform computational tasks with superior power efficiency to conventional computers. Approaches that use traditional electronic devices to create artificial neurons and synapses are, however, currently limited by the energy and area requirements of these components. Spintronic nanodevices, which exploit both the magnetic and electrical properties of electrons, can increase the energy efficiency and decrease the area of these circuits, and magnetic tunnel junctions are of particular interest as neuromorphic computing elements because they are compatible with standard integrated circuits and can support multiple functionalities. Here we review the development of spintronic devices for neuromorphic computing. We examine how magnetic tunnel junctions can serve as synapses and neurons, and how magnetic textures, such as domain walls and skyrmions, can function as neurons. We also explore spintronics-based implementations of neuromorphic computing tasks, such as pattern recognition in an associative memory, and discuss the challenges that exist in scaling up these systems.

physics.app-ph↗

arXiv2019

Temporal pattern recognition with delayed feedback spin-torque nano-oscillators

The recent demonstration of neuromorphic computing with spin-torque nano-oscillators has opened a path to energy efficient data processing. The success of this demonstration hinged on the intrinsic short-term memory of the oscillators. In this study, we extend the memory of the spin-torque nano-oscillators through time-delayed feedback. We leverage this extrinsic memory to increase the efficiency of solving pattern recognition tasks that require memory to discriminate different inputs. The large tunability of these non-linear oscillators allows us to control and optimize the delayed feedback memory using different operating conditions of applied current and magnetic field.

physics.app-ph↗

arXiv2019

Enhancing the injection locking range of spin torque oscillators through mutual coupling

We investigate how the ability of the vortex oscillation mode of a spin-torque nano-oscillator to lock to an external microwave signal is modified when it is coupled to another oscillator. We show experimentally that mutual electrical coupling can lead to locking range enhancements of a factor 1.64. Furthermore, we analyze the evolution of the locking range as a function of the coupling strength through experiments and numerical simulations. By uncovering the mechanisms at stake in the locking range enhancement, our results will be useful for designing spin-torque nano-oscillators arrays with high sensitivities to external microwave stimuli.

cond-mat.mes-hall↗

arXiv2019

Neuromorphic Computing through Time-Multiplexing with a Spin-Torque Nano-Oscillator

Fabricating powerful neuromorphic chips the size of a thumb requires miniaturizing their basic units: synapses and neurons. The challenge for neurons is to scale them down to submicrometer diameters while maintaining the properties that allow for reliable information processing: high signal to noise ratio, endurance, stability, reproducibility. In this work, we show that compact spin-torque nano-oscillators can naturally implement such neurons, and quantify their ability to realize an actual cognitive task. In particular, we show that they can naturally implement reservoir computing with high performance and detail the recipes for this capability.

cs.ET↗

arXiv2019

Microwave neural processing and broadcasting with spintronic nano-oscillators

Can we build small neuromorphic chips capable of training deep networks with billions of parameters? This challenge requires hardware neurons and synapses with nanometric dimensions, which can be individually tuned, and densely connected. While nanosynaptic devices have been pursued actively in recent years, much less has been done on nanoscale artificial neurons. In this paper, we show that spintronic nano-oscillators are promising to implement analog hardware neurons that can be densely interconnected through electromagnetic signals. We show how spintronic oscillators maps the requirements of artificial neurons. We then show experimentally how an ensemble of four coupled oscillators can learn to classify all twelve American vowels, realizing the most complicated tasks performed by nanoscale neurons.

physics.app-ph↗

arXiv2016

Driven energy transfer between coupled modes in spin-torque oscillators

The mutual interaction between the different eigenmodes of a spin-torque oscillator can lead to a large variety of physical mechanisms from mode hopping to multi-mode generation, that usually reduce their performances as radio-frequency devices. To tackle this issue for the future applications, we investigate the properties of a model spin-torque oscillator that is composed of two coupled vortices with one vortex in each of the two magnetic layers of the oscillator. In such double-vortex system, the remarkable properties of energy transfer between the coupled modes, one being excited by spin transfer torque while the second one being damped, result into an alteration of the damping parameters. As a consequence, the oscillator nonlinear behavior is concomitantly drastically impacted. This efficient coupling mechanism, driven mainly by the dynamic dipolar field generated by the spin transfer torque induced motion of the vortices, gives rise to an unexpected dynamical regime of self-resonance excitation. These results show that mode coupling can be leveraged for controlling the synchronization process as well as the frequency tunability of spin-torque oscillators.

cond-mat.mes-hall↗

arXiv2016

Synchronization of electrically coupled stochastic magnetic oscillators induced by thermal and electrical noise

Superparamagnetic tunnel junctions are nanostructures that auto-oscillate stochastically under the effect of thermal noise. Recent works showed that despite their stochasticity, such junctions possess a capability to synchronize to subthreshold voltage drives, in a way that can be enhanced or controlled by adding noise. In this work, we investigate a system composed of two electrically coupled junctions, connected in series to a periodic voltage source. We make use of numerical simulations and of an analytical model to demonstrate that both junctions can be phase-locked to the drive, in phase or in anti-phase. This synchronization phenomenon can be controlled by both thermal and electrical noises, although the two types of noises induce qualitatively different behaviors. Namely, thermal noise can stabilize a regime where one junction is phase-locked to the drive voltage while the other is blocked in one state. On the contrary, electrical noise causes the junctions to have highly correlated behaviors and thus cannot induce the latter. These results open the way for the design of superparamagnetic tunnel junctions that can perform computation through synchronization, and which harvest the largest part of their energy consumption from thermal noise.

cond-mat.mes-hall↗

arXiv2016

Controlling the phase locking of unstable magnetic bits for ultra-low power computation

When fabricating magnetic memories, one of the main challenges is to maintain the bit stability while downscaling. Indeed, for magnetic volumes of a few thousand nm3, the energy barrier between magnetic configurations becomes comparable to the thermal energy at room temperature. Then, switches of the magnetization spontaneously occur. These volatile, superparamagnetic nanomagnets are generally considered useless. But what if we could use them as low power computational building blocks? Remarkably, they can oscillate without the need of any external dc drive, and despite their stochastic nature, they can beat in unison with an external periodic signal. Here we show that the phase locking of superparamagnetic tunnel junctions can be induced and suppressed by electrical noise injection. We develop a comprehensive model giving the conditions for synchronization, and predict that it can be achieved with a total energy cost lower than 10-^13 J. Our results open the path to ultra-low power computation based on the controlled synchronization of oscillators.

cond-mat.mtrl-sci↗

arXiv2016

Mutual synchronization of spin torque nano-oscillators through a non-local and tunable electrical coupling

The concept of spin torque driven high frequency magnetization dynamics has opened up the field of spintronics to non-linear physics, potentially in complex networks of dynamical systems. In the scarce demonstrations of synchronized spin-torque oscillators, the local nature of the magnetic coupling that is used have largely hampered a good understanding and thus the control of the synchronization process. Here we take advantage of the non-local nature of an electrical coupling to mutually synchronize spin-torque oscillators through their self-emitted microwave currents. The control of the synchronized state is achieved at the nanoscale through two active components of spin transfer torques, but also externally through an electrical delay line. These additional levels of control of the synchronization capability provide new approaches to underlie a large variety of nanoscale collective dynamics in complex networks.

cond-mat.mes-hall↗

arXiv2015

Efficient Synchronization of Dipolarly Coupled Vortex-Based Spin Transfer Nano-Oscillators

Due to their nonlinear properties, spin transfer nano-oscillators can easily adapt their frequency to external stimuli. This makes them interesting model systems to study the effects of synchronization and brings some opportunities to improve their microwave characteristics in view of their applications in information and communication technologies and to design innovative computing architectures. So far, mutual synchronization of spin transfer nano-oscillators through propagating spin-waves and exchange coupling in a common magnetic layer has been demonstrated. Here we show that the dipolar interaction is also an efficient mechanism to synchronize neighbouring oscillators. We experimentally study a pair of vortex-based spin-transfer nano-oscillators, in which mutual synchronization can be achieved despite a significant frequency mismatch between oscillators. Importantly, the coupling efficiency is controlled by the magnetic configuration of the vortices, as confirmed by an analytical model highlighting the physics at play in the synchronization process as well as by micromagnetic simulations.

cond-mat.mes-hall↗

arXiv2015

Self-injection locking of a vortex spin torque oscillator by delayed feedback

The self-synchronization of spin torque oscillators is investigated experimentally by re-injecting its radiofrequency (rf) current after a certain delay time. We demonstrate that the emission power and the spectral linewidth are improved for optimal delay times. Moreover by varying the phase difference between the emitted power and the re-injected one, we find a clear oscillatory dependence with a 2π periodicity of the frequency of the oscillator as well as its power and linewidth. Such periodical behavior within the self-injection regime is well described by the general model of nonlinear auto-oscillators including not only a delayed rf current but also all spin torque forces responsible for the self-synchronization. Our results reveal new approaches for controlling the non-autonomous dynamics of spin torque oscillators, a key issue for rf spintronics applications as well as for the development of neuro-inspired spin-torque oscillators based devices.

cond-mat.mes-hall↗

arXiv2015

Optimizing magneto-dipolar interactions for synchronizing vortex based spin-torque nano-oscillators

We report on a theoretical study about the magneto-dipolar coupling and synchronization between two vortex-based spin-torque nano-oscillators. In this work we study the dependence of the coupling efficiency on the relative magnetization parameters of the vortices in the system. For that purpose, we combine micromagnetic simulations, Thiele equation approach, and analytical macro-dipole approximation model to identify the optimized configuration for achieving phase-locking between neighboring oscillators. Notably, we compare vortices configurations with parallel (P) polarities and with opposite (AP) polarities. We demonstrate that the AP core configuration exhibits a coupling strength about three times larger than in the P core configuration.

cond-mat.mes-hall↗

arXiv2015

Critical current and linewidth reduction in spin-torque nano-oscillators by delayed self-injection

Based on theoretical models, the dynamics of spin-torque nano-oscillators can be substantially modified by re-injecting the emitted signal to the input of the oscillator after some delay. Numerical simulations for vortex magnetic tunnel junctions show that with reasonable parameters this approach can decrease critical currents as much as 25 % and linewidths by a factor of 4. Analytical calculations, which agree well with simulations, demonstrate that these results can be generalized to any kind of spin-torque oscillator.

cond-mat.mes-hall↗

arXiv2015

Understanding of phase noise squeezing under fractional synchronization of non-linear spin transfer vortex oscillator

We investigate experimentally the synchronization of a vortex based spin transfer oscillator to an external rf current whose frequency is at multiple integers, as well as half integer, of the oscillator frequency. Through a theoretical study of the locking process, we highlight both the crucial role of the symmetries of the spin torques acting on the magnetic vortex and the nonlinear properties of the oscillator on the phase locking process. Through the achievement of a perfect injection locking state, we report a record phase noise reduction down to -90dBc/Hz at 1 kHz offset frequency. The phase noise of these nanoscale oscillators is demonstrating as being low and controllable which is of significant importance for real applications using spin transfer devices.

cond-mat.mes-hall↗

arXiv2014

Spin transfer torques generated by the anomalous Hall effect and anisotropic magnetoresistance

Spin-orbit coupling in ferromagnets gives rise to the anomalous Hall effect and the anisotropic magnetoresistance, both of which can be used to create spin-transfer torques in a similar manner as the spin Hall effect. In this paper we show how these effects can be used to reliably switch perpendicularly magnetized layers and to move domain walls. A drift-diffusion treatment of the anomalous Hall effect and the anisotropic magnetoresistance describes the spin currents that flow in directions perpendicular to the electric field. In systems with two ferromagnetic layers separated by a spacer layer, an in-plane electric field cause spin currents to be injected from one layer into the other, creating spin transfer torques. Unlike the related spin Hall effect in non-magnetic materials, the anomalous Hall effect and the anisotropic magnetoresistance allow control of the orientation of the injected spins, and hence torques, by changing the direction of the magnetization in the injecting layer. The torques on one layer show a rich angular dependence as a function of the orientation of the magnetization in the other layer. The control of the torques afforded by changing the orientation of the magnetization in a fixed layer makes it possible to reliably switch a perpendicularly magnetized free layer. Our calculated critical current densities for a representative CoFe/Cu/FePt structure show that the switching can be efficient for appropriate material choices. Similarly, control of the magnetization direction can drive domain wall motion, as shown for NiFe/Cu/NiFe structures.

cond-mat.mtrl-sci↗