Search

Charge-spin interconversion processes underpin the generation of spin-orbit torques in magnetic/nonmagnetic bilayers. However, efficient sources of spin currents such as 5d metals are also efficient spin sinks, resulting in a large increase of magnetic damping. Here we show that a partially-oxidized 3d metal can generate a strong orbital torque without a significant increase in damping. Measurements of the torque efficiency {\xi} and Gilbert damping {\alpha} in CoFe/CuOx and CoFe/Pt indicate that {\xi} is comparable. The increase in damping relative to a single CoFe layer is {\Delta}{\alpha}<0.002 in CoFe/CuOx and {\Delta}{\alpha} ~ 0.005 - 0.02 in CoFe/Pt, depending on CoFe thickness. We ascribe the nonreciprocal relationship between {\Delta}{\alpha} and {\xi} in CoFe/CuOx to the small orbital-to-spin current ratio generated by magnetic resonance in CoFe and the lack of an efficient spin sink in CuOx. Our findings provide new perspectives on the efficient excitation of magnetization dynamics via the orbital torque.

The magnetic resonance of iron garnets is at the heart of rich physics and provides a crucial technology used inside microwave components. A barrier still stands in the way of integrated cryogenic devices with iron garnets, because their epitaxy requires paramagnetic substrates that cause large microwave losses at low temperatures. Finding alternative combinations of substrates and magnetic garnets is thus essential to enabling cryogenic integrated magnonics. With this motivation, we develop a new approach to the synthesis of epitaxial garnet films, based on the co-sputtering of binary oxides, enabling a precise and continuous tuning of the iron garnet composition. By substituting a controlled proportion of the iron with additional yttrium, we obtain Y$_{3}$(Y$_{x}$Fe$_{5-x}$)O$_{12}$ films of high crystalline quality that can be lattice-matched to yttrium scandium gallium garnet (YSGG), a diamagnetic substrate suitable for low-temperature applications. Ferromagnetic resonance performed at low temperatures confirms the elimination of substrate paramagnetism losses on YSGG. We also show that the Y$_{3}$(Y$_{x}$Fe$_{5-x}$)O$_{12}$ system can be matched to other substrates, such as gadolinium yttrium scandium gallium garnet and gadolinium scandium gallium garnet. Additionally, we demonstrate that Bi-substituted films of (Bi$_{0.8}$Y$_{2.2}$)Fe$_{5}$O$_{12}$ also offer ideal lattice matching to YSGG for low-temperature magnetic resonance. The present approach of controlled substitutions in iron garnets to obtain ideal lattice matching with diamagnetic substrates opens a very promising pathway to magnonic devices operating in low-temperature environments., Comment: Main: 18 pages, 6 figures; Supplement: 9 pages, 6 figures

Orbital currents have recently emerged as a promising tool to achieve electrical control of the magnetization in thin-film ferromagnets. Efficient orbital-to-spin conversion is required in order to torque the magnetization. Here we show that the injection of an orbital current in a ferrimagnetic GdyCo100-y alloy generates strong orbital torques whose sign and magnitude can be tuned by changing the Gd content and temperature. The effective spin-orbital Hall angle reaches up to -0.25 in a GdyCo100-y/CuOx bilayer compared to +0.03 in Co/CuOx and +0.13 in GdyCo100-y/Pt. This behavior is attributed to the local orbital-to-spin conversion taking place at the Gd sites, which is about five times stronger and of the opposite sign relative to Co. Furthermore, we observe a manyfold increase in the net orbital torque at low temperature, which we attribute to the improved conversion efficiency following the magnetic ordering of the Gd and Co sublattices.

Control over quantum systems is typically achieved by time-dependent electric or magnetic fields. Alternatively, electronic spins can be controlled by spin-polarized currents. Here we demonstrate coherent driving of a single spin by a radiofrequency spin-polarized current injected from the tip of a scanning tunneling microscope into an organic molecule. With the excitation of electron paramagnetic resonance, we established dynamic control of single spins by spin torque using a local electric current. In addition our work highlights the dissipative action of the spin-transfer torque, in contrast to the nondissipative action of the magnetic field, which allows for the manipulation of individual spins based on controlled decoherence.

The ability (and inability) of large language models (LLMs) to perform arithmetic tasks has been the subject of much theoretical and practical debate. We show that LLMs are frequently able to correctly and confidently predict the first digit of n-digit by m-digit multiplication tasks without using chain of thought reasoning, despite these tasks require compounding operations to solve. Simultaneously, LLMs in practice often fail to correctly or confidently predict the last digit of an n-digit by m-digit multiplication, a task equivalent to 1-digit by 1-digit multiplication which can be easily learned or memorized. We show that the latter task can be solved more robustly when the LLM is conditioned on all of the correct higher-order digits, which on average increases the confidence of the correct last digit on 5-digit by 5-digit multiplication tasks using Llama 2-13B by over 230% (0.13 to 0.43) and Mistral-7B by 150% (0.22 to 0.55)., Comment: In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

In this work, we compute the lower bound of the integrality gap of the Metric Steiner Tree Problem (MSTP) on a graph for some small values of number of nodes and terminals. After debating about some limitations of the most used formulation for the Steiner Tree Problem, namely the Bidirected Cut Formulation, we introduce a novel formulation, that we named Complete Metric formulation, tailored for the metric case. We prove some interesting properties of this formulation and characterize some types of vertices. Finally, we define a linear program (LP) by adapting a method already used in the context of the Travelling Salesman Problem. This LP takes as input a vertex of the polytope of the CM relaxation and provides an MSTP instance such that (a) the optimal solution is precisely that vertex and (b) among all of the instances having that vertex as its optimal solution, the selected instance is the one having the highest integrality gap. We propose two heuristics for generating vertices to provide inputs for our procedure. In conclusion, we raise several conjectures and open questions., Comment: 39 pages, 5 figures

Generative models, e.g., Stable Diffusion, have enabled the creation of photorealistic images from text prompts. Yet, the generation of 360-degree panorama images from text remains a challenge, particularly due to the dearth of paired text-panorama data and the domain gap between panorama and perspective images. In this paper, we introduce a novel dual-branch diffusion model named PanFusion to generate a 360-degree image from a text prompt. We leverage the stable diffusion model as one branch to provide prior knowledge in natural image generation and register it to another panorama branch for holistic image generation. We propose a unique cross-attention mechanism with projection awareness to minimize distortion during the collaborative denoising process. Our experiments validate that PanFusion surpasses existing methods and, thanks to its dual-branch structure, can integrate additional constraints like room layout for customized panorama outputs. Code is available at https://chengzhag.github.io/publication/panfusion., Comment: CVPR 2024. Project Page: https://chengzhag.github.io/publication/panfusion Code: https://github.com/chengzhag/PanFusion

The field of computational thinking education has grown in recent years as researchers and educators have sought to develop and assess students' computational thinking abilities. While much of the research in this area has focused on defining computational thinking, the competencies it involves and how to assess them in teaching and learning contexts, this work takes a different approach. We provide a more situated perspective on computational thinking, focusing on the types of problems that require computational thinking skills to be solved and the features that support these processes. We develop a framework for analysing existing computational thinking problems in an educational context. We conduct a comprehensive literature review to identify prototypical activities from areas where computational thinking is typically pursued in education. We identify the main components and characteristics of these activities, along with their influence on activating computational thinking competencies. The framework provides a catalogue of computational thinking skills that can be used to understand the relationship between problem features and competencies activated. This study contributes to the field of computational thinking education by offering a tool for evaluating and revising existing problems to activate specific skills and for assisting in designing new problems that target the development of particular competencies. The results of this study may be of interest to researchers and educators working in computational thinking education.

We report an axion dark matter search with a haloscope equipped with a microwave photon counter. The haloscope is a tunable high quality factor 3-dimensional microwave cavity placed in a magnetic field. The photon counter, operated cyclically, maps an incoming microwave photon onto the state of a superconducting transmon qubit. The measurement protocol continuously monitors the power emitted by the haloscope cavity as well as the dark count background, and enables tuning of the cavity frequency to probe different axion masses. With this apparatus we enhance by a factor 20 the search speed that can be reached with quantum-limited linear amplifiers, and set a new standard for probing the existence of axions with resonant detectors.

We report the first experimental results obtained with the new haloscope of the QUAX experiment located at Laboratori Nazionali di Frascati of INFN (LNF). The haloscope is composed of a OFHC Cu resonant cavity cooled down to about 30 mK and immersed in a magnetic field of 8 T. The cavity frequency was varied in a 6 MHz range between 8.831496 and 8.83803 GHz. This corresponds to a previously unprobed mass range between 36.52413 and 36.5511 $\mu$eV. We don't observe any excess in the power spectrum and set limits on the axion-photon coupling in this mass range down to $g_{a\gamma\gamma} < 0.861 \times 10^{-13}$ GeV$^{-1}$ with the confidence level set at $90\%$., Comment: Submitted to Physical Review D (https://journals.aps.org/prd/)

Recent developments in foundation models, like Large Language Models (LLMs) and Vision-Language Models (VLMs), trained on extensive data, facilitate flexible application across different tasks and modalities. Their impact spans various fields, including healthcare, education, and robotics. This paper provides an overview of the practical application of foundation models in real-world robotics, with a primary emphasis on the replacement of specific components within existing robot systems. The summary encompasses the perspective of input-output relationships in foundation models, as well as their role in perception, motion planning, and control within the field of robotics. This paper concludes with a discussion of future challenges and implications for practical robot applications.

We study the incubation and transition times that characterize the magnetization switching induced by spin-orbit torques in nanomagnets with perpendicular anisotropy. We present a phenomenological model to interpret the dependence of the incubation time on the amplitude of the voltage pulse and assisting magnetic field and estimate the volume of the seed domain that triggers the switching. Our measurements evidence a correlation between the incubation and transition times that is mediated by the temperature variation during the electric pulse. In addition, we discuss the stochastic distributions of the two times in terms of the energy barriers opposing the nucleation and expansion of the seed domain. We propose two models based on the log-normal and gamma functions to account for the different origin of the variability of the incubation and transition times, which are associated with a single nucleation barrier and multiple pinning sites, respectively.

The Hanle magnetoresistance is a telltale signature of spin precession in nonmagnetic conductors, in which strong spin-orbit coupling generates edge spin accumulation via the spin Hall effect. Here, we report the existence of a large Hanle magnetoresistance in single layers of Mn with weak spin-orbit coupling, which we attribute to the orbital Hall effect. The simultaneous observation of a sizable Hanle magnetoresistance and vanishing small spin Hall magnetoresistance in BiYIG/Mn bilayers corroborates the orbital origin of both effects. We estimate an orbital Hall angle of 0.016, an orbital relaxation time of 2 ps and diffusion length of the order of 2 nm in disordered Mn. Our findings indicate that current-induced orbital moments are responsible for magnetoresistance effects comparable to or even larger than those determined by spin moments, and provide a tool to investigate nonequilibrium orbital transport phenomena.

Light transition metals have recently emerged as a sustainable material class for efficient spin-charge interconversion. We report measurements of current-induced spin-orbit torques generated by Ni$_{1-x}$Cu$_{x}$ alloys in perpendicularly magnetized ferrimagnetic Gd/Fe multilayers. We show that the spin-orbit torque efficiency of Ni$_{1-x}$Cu$_{x}$ increases with the Ni/Cu atomic ratio, reaching values comparable to those of Pt for Ni$_{55}$Cu$_{45}$. Furthermore, we demonstrate magnetization switching of a 20-nm-thick Gd/Fe multilayer with a threshold current that decreases with increasing Ni concentration, similar to the spin-orbit torque efficiency. Our findings show that Ni$_{1-x}$Cu$_{x}$$-$based magnetic heterostructures allow for efficient control of the magnetization by electric currents.

We explore the reach of analytical models at one-loop in Perturbation Theory (PT) to accurately describe measurements of the galaxy power spectrum from numerical simulations in redshift space. We consider the validity range in terms of three different diagnostics: 1) the goodness of fit; 2) a figure-of-bias quantifying the error in recovering the fiducial value of a cosmological parameter; 3) an internal consistency check of the theoretical model quantifying the running of the model parameters with the scale cut. We consider different sets of measurements corresponding to an increasing cumulative simulation volume in redshift space. For each volume we define a median value and the associated scatter for the largest wavenumber where the model is valid (the $k$-reach of the model). We find, as a rather general result, that the median value of the reach decreases with the simulation volume, as expected since the smaller statistical errors provide a more stringent test for the model. This is true for all the three definitions considered, with the one given in terms of the figure-of-bias providing the most stringent scale cut. More interestingly, we find as well that the error associated with the $k$-reach value is quite large, with a significant probability of being as low as 0.1$\, h \, {\rm Mpc}^{-1}$ (or, more generally, up to 40% smaller than the median) for all the simulation volumes considered. We explore as well the additional information on the growth rate parameter encoded in the power spectrum hexadecapole, compared to the analysis of monopole and quadrupole, as a function of simulation volume. While our analysis is, in many ways, rather simplified, we find that the gain in the determination of the growth rate is quite small in absolute value and well within the statistical error on the corresponding figure of merit., Comment: 7 pages, 3 figures

Topological insulators have attracted great interest as generators of spin-orbit torques (SOTs) in spintronic devices. Bi\textsubscript{1-x}Sb\textsubscript{x} is a prominent topological insulator that has a high charge-to-spin conversion efficiency. However, the origin and magnitude of the SOTs induced by current-injection in Bi\textsubscript{1-x}Sb\textsubscript{x} remain controversial. Here we report the investigation of the SOTs and spin Hall magnetoresistance resulting from charge-to-spin conversion in twin-free epitaxial layers of Bi\textsubscript{0.9}Sb\textsubscript{0.1}(0001) coupled to FeCo, and compare it with that of amorphous Bi\textsubscript{0.9}Sb\textsubscript{0.1}. We find a large charge-to-spin conversion efficiency of 1 in the first case and less than 0.1 in the second, confirming crystalline Bi\textsubscript{0.9}Sb\textsubscript{0.1} as a strong spin injector material. The SOTs and spin Hall magnetoresistance are independent of the direction of the electric current, indicating that charge-to-spin conversion in single-crystal Bi\textsubscript{0.9}Sb\textsubscript{0.1}(0001) is isotropic despite the strong anisotropy of the topological surface states. Further, we find that the damping-like SOT has a non-monotonic temperature dependence with a minimum at 20~K. By correlating the SOT with resistivity and weak antilocalization measurements, we conclude that charge-spin conversion occurs via thermally-excited holes from the bulk states above 20~K, and conduction through the isotropic surface states with increasing spin polarization due to decreasing electron-electron scattering below 20~K., Comment: Supporting information at https://doi.org/10.1002/adma.202304905

In this article, we prove a Reocurrence Theorem over function fields of curves over $\mathbf{C}(\! (t)\! )$ and over finite extensions of the Laurent series field $\mathbf{C}(\! (x,y)\! )$. This provides a partial replacement to Chebotarev's Theorem over such fields. A concrete application to the study of weak approximation for homogeneous spaces under $\mathrm{SL}_n$ and with finite stabilizers is given at the end of the article.

X-ray absorption spectroscopy (XAS) and x-ray magnetic circular dichroism (XMCD) have been used to study transition metal impurities on K and Na films. The multiplet structure of the XAS spectra indicates that Fe, Co, and Ni have localized atomic ground states with predominantly d7, d8, and d9 character, respectively. XMCD shows that the localized impurity states possess large, atomiclike, magnetic orbital moments that are progressively quenched as clusters are formed. Ni impurities on Na films are found to be nonmagnetic, with a strongly increased d10 character of the impurity state. The results show that the high magnetic moments of transition metals in alkali hosts originate from electron localization.

This paper discusses a heuristic approach for Team Orienteering Problems with Time Windows. The method we propose takes advantage of a solution model based on a hierarchic generalization of the original problem, which is combined with an Ant Colony System algorithm. Computational results on benchmark instances previously adopted in the literature suggest that the algorithm we propose is effective in practice.

In creative design, where aesthetics play a crucial role in determining the quality of outcomes, there are often multiple worthwhile possibilities, rather than a single ``best'' design. This challenge is compounded in the use of computational generative systems, where the sheer number of potential outcomes can be overwhelming. This paper introduces a method that combines evolutionary optimisation with AI-based image classification to perform quality-diversity search, allowing for the creative exploration of complex design spaces. The process begins by randomly sampling the genotype space, followed by mapping the generated phenotypes to a reduced representation of the solution space, as well as evaluating them based on their visual characteristics. This results in an elite group of diverse outcomes that span the solution space. The elite is then progressively updated via sampling and simple mutation. We tested our method on a generative system that produces abstract drawings. The results demonstrate that the system can effectively evolve populations of phenotypes with high aesthetic value and greater visual diversity compared to traditional optimisation-focused evolutionary approaches., Comment: Paper accepted to GECCO 2023 Conference

Current-induced spin-orbit torques (SOTs) enable fast and efficient manipulation of the magnetic state of magnetic tunnel junctions (MTJs), making it attractive for memory, in-memory computing, and logic applications. However, the requirement of the external magnetic field to achieve deterministic switching in perpendicular magnetized SOT-MTJs limits its implementation for practical applications. Here, we introduce a field-free switching (FFS) solution for the SOT-MTJ device by shaping the SOT channel to create a "bend" in the SOT current. The resulting bend in the charge current creates a spatially non-uniform spin current, which translates into inhomogeneous SOT on an adjacent magnetic free layer enabling deterministic switching. We demonstrate FFS experimentally on scaled SOT-MTJs at nanosecond time scales. This proposed scheme is scalable, material-agnostic, and readily compatible with wafer-scale manufacturing, thus creating a pathway for developing purely current-driven SOT systems.

Two-dimensional arrays of magnetically coupled nanomagnets provide a mesoscopic platform for exploring collective phenomena as well as realizing a broad range of spintronic devices. In particular, the magnetic coupling plays a critical role in determining the nature of the cooperative behaviour and providing new functionalities in nanomagnet-based devices. Here, we create coupled Ising-like nanomagnets in which the coupling between adjacent nanomagnetic regions can be reversibly converted between parallel and antiparallel through solid-state ionic gating. This is achieved with the voltage-control of magnetic anisotropies in a nanosized region where the symmetric exchange interaction favours parallel alignment and the antisymmetric exchange interaction, namely the Dzyaloshinskii-Moriya interaction, favours antiparallel alignment. Applying this concept to a two-dimensional lattice, we demonstrate a voltage-controlled phase transition in artificial spin ices. Furthermore, we achieve an addressable control of the individual couplings and realize an electrically programmable Ising network, which opens up new avenues to design nanomagnet-based logic devices and neuromorphic computers

A traveling wave parametric amplifier has been integrated in the haloscope of the QUAX experiment. A search for dark matter axions has been performed with a high Q dielectric cavity immersed in a 8 T magnetic field and read by a detection chain having a system noise temperature of about 2.1 K at the frequency of 10.353 GHz. Scanning has been conducted by varying the cavity frequency using sapphire rods immersed into the cavity. At multiple operating frequencies, the sensitivity of the instrument was at the level of viable axion models.

Strong, adjustable magnetic couplings are of great importance to all devices based on magnetic materials. Controlling the coupling between adjacent regions of a single magnetic layer, however, is challenging. In this work, we demonstrate strong exchange-based coupling between arbitrarily shaped regions of a single ferrimagnetic layer. This is achieved by spatially patterning the compensation temperature of the ferrimagnet by either oxidation or He+ irradiation. The coupling originates at the lateral interface between regions with different compensation temperature and scales inversely with their width. We show that this coupling generates large lateral exchange coupling fields and we demonstrate its application to control the switching of magnetically compensated dots with an electric current.

In recent years Generative Machine Learning systems have advanced significantly. A current wave of generative systems use text prompts to create complex imagery, video, even 3D datasets. The creators of these systems claim a revolution in bringing creativity and art to anyone who can type a prompt. In this position paper, we question the basis for these claims, dividing our analysis into three areas: the limitations of linguistic descriptions, implications of the dataset, and lastly, matters of materiality and embodiment. We conclude with an analysis of the creative possibilities enabled by prompt-based systems, asking if they can be considered a new artistic medium., Comment: Paper accepted for EvoMUSART Conference, Brno, Czech Republic, 12-14 April 2023

Magnetic skyrmions are compact chiral spin textures that exhibit a rich variety of topological phenomena and hold potential for developing high-density memory devices and novel computing schemes driven by spin currents. Here, we demonstrate room temperature interfacial stabilization and current-driven control of skyrmion bubbles in the ferrimagnetic insulator Tm3Fe5O12 (TmIG) coupled to Pt. We track the current-induced motion of individual skyrmion bubbles. The ferrimagnetic order of the crystal together with the interplay of spin-orbit torques and pinning determine the skyrmion dynamics in TmIG and result in a strong skyrmion Hall effect characterized by a negative deflection angle and hopping motion. Further, we show that the velocity and depinning threshold of the skyrmion bubbles can be modified by exchange coupling TmIG to an in-plane magnetized Y3Fe5O12 layer, which distorts the spin texture of the skyrmions and leads to a directional-dependent rectification of their dynamics. This effect, which is equivalent to a magnetic ratchet, is exploited to control the skyrmion flow in a racetrack-like device., Comment: 52 pages, 24 figures (manuscript, extended data and supplementary information)

We introduce an efficient algorithm for general data mosaicing, based on the simulation-based inference paradigm. Our algorithm takes as input a target datum, source data, and partitions of the target and source data into fragments, learning distributions over averages of fragments of the source data such that samples from those distributions approximate fragments of the target datum. We utilize a model that can be trivially parallelized in conjunction with the latest advances in efficient simulation-based inference in order to find approximate posteriors fast enough for use in practical applications. We demonstrate our technique is effective in both audio and image mosaicing problems.

We consider the task of detecting anomalies for autonomous mobile robots based on vision. We categorize relevant types of visual anomalies and discuss how they can be detected by unsupervised deep learning methods. We propose a novel dataset built specifically for this task, on which we test a state-of-the-art approach; we finally discuss deployment in a real scenario., Comment: Workshop paper presented at the ICRA 2022 Workshop on Safe and Reliable Robot Autonomy under Uncertainty https://sites.google.com/umich.edu/saferobotautonomy/home

We consider the problem of building visual anomaly detection systems for mobile robots. Standard anomaly detection models are trained using large datasets composed only of non-anomalous data. However, in robotics applications, it is often the case that (potentially very few) examples of anomalies are available. We tackle the problem of exploiting these data to improve the performance of a Real-NVP anomaly detection model, by minimizing, jointly with the Real-NVP loss, an auxiliary outlier exposure margin loss. We perform quantitative experiments on a novel dataset (which we publish as supplementary material) designed for anomaly detection in an indoor patrolling scenario. On a disjoint test set, our approach outperforms alternatives and shows that exposing even a small number of anomalous frames yields significant performance improvements.

We report the observation of a unidirectional magnetoresistance (UMR) that originates from the nonequilibrium orbital momentum induced by an electric current in a naturally oxidized Cu/Co bilayer. The orbital-UMR scales with the torque efficiency due to the orbital Rashba-Edelstein effect upon changing the Co thickness and temperature, reflecting their common origin. We attribute the UMR to orbital-dependent electron scattering and orbital-to-spin conversion in the ferromagnetic layer. In contrast to the spin-current induced UMR, the magnon contribution to the orbital-UMR is absent in thin Co layers, which we ascribe to the lack of coupling between low energy magnons and orbital current. The magnon contribution to the UMR emerges in Co layers thicker than about 5 nm, which is comparable to the orbital-to-spin conversion length. Our results provide insight into orbital-to-spin momentum transfer processes relevant for the optimization of spintronic devices based on light metals and orbital transport., Comment: 12 pages, 3 figures

A haloscope of the QUAX--$a\gamma$ experiment, composed of an high-Q resonant cavity immersed in a 8 T magnet and cooled to $\sim 4.5$~K is operated to search for galactic axion with mass $m_a\simeq42.8~\mu\text{eV}$. The design of the cavity with hollow dielectric cylinders concentrically inserted in a OFHC Cu cavity, allowed us to maintain a loaded quality-factor Q $\sim 300000$ during the measurements in presence of magnetic field. Through the cavity tuning mechanism it was possible to modulate the resonance frequency of the haloscope in the region $10.35337-10.35345$~GHz and thus acquire different dataset at different resonance frequencies. Acquiring each dataset for about 50 minutes, combining them and correcting for the axion's signal estimation-efficiency we set a limit on the axion-photon coupling $g_{a\gamma\gamma}< 0.731\times10^{-13}$ GeV$^{-1}$ with the confidence level set at $90\%$.

Spin-orbit torques (SOT) provide a versatile tool to manipulate the magnetization of diverse classes of materials and devices using electric currents, leading to novel spintronic memory and computing approaches. In parallel to spin transfer torques (STT), which have emerged as a leading non-volatile memory technologie, SOT broaden the scope of current-induced magnetic switching to applications that run close to the clock speed of the central processing unit and unconventional computing architectures. In this paper, we review the fundamental characteristics of SOT and their use to switch magnetic tunnel junction (MTJ) devices, the elementary unit of the magnetoresistive random access memory (MRAM). In the first part, we illustrate the physical mechanisms that drive the SOT and magnetization reversal in nanoscale structures. In the second part, we focus on the SOT-MTJ cell. We discuss the anatomy of the MTJ in terms of materials and stack development, summarize the figures of merit for SOT switching, review the field-free operation of perpendicularly magnetized MTJs, and present options to combine SOT, STT and voltage-gate assisted switching. In the third part, we consider SOT-MRAMs in the perspective of circuit integration processes, introducing considerations on scaling and performance, as well as macro-design architectures. We thus bridge the fundamental description of SOT-driven magnetization dynamics with an application-oriented perspective, including device and system-level considerations, goals, and challenges.

The orbital Hall effect provides an alternative means to the spin Hall effect to convert a charge current into a flow of angular momentum. Recently, compelling signatures of orbital Hall effects have been identified in 3d transition metals. Here, we report a systematic study of the generation, transmission, and conversion of orbital currents in heterostructures comprising 3d, 5d, and 4f metals. We show that the orbital Hall conductivity of Cr reaches giant values of the order of 5*10^5 Ohm^{-1} m^{-1} and that Pt presents a strong orbital Hall effect in addition to the spin Hall effect. Measurements performed as a function of thickness of nonmagnetic Cr, Mn, and Pt layers and ferromagnetic Co and Ni layers reveal how the orbital and spin currents compete or assist each other in determining the spin-orbit torques acting on the magnetic layer. We further show how this interplay can be drastically modulated by introducing 4 f spacers between the nonmagnetic and magnetic layers. Gd and Tb act as very efficient orbital-to-spin current converters, boosting the spin-orbit torques generated by Cr by a factor of 4 and reversing the sign of the torques generated by Pt. To interpret our results, we present a generalized drift-diffusion model that includes both spin and orbital Hall effects and describes their interconversion mediated by spin-orbit coupling.

We investigate the motion of magnetic domain walls driven by magnetic fields and current-driven spin-orbit torques in an exchange-biased system with perpendicular magnetization. We consider Cr2O3/Co/Pt trilayers as model system, in which the magnetization of the Co layer can be exchanged-biased out-of-plane or in-plane depending on the field cooling direction. In field-driven experiments, the in-plane exchange bias favors the propagation of the domain walls with internal magnetization parallel to the exchange bias field. In current-driven experiments, the domain walls propagate along the current direction, but the domain wall velocity increases and decreases symmetrically (anti-symmetrically) for both current polarities when the exchange bias is parallel (perpendicular) to the current line. At zero external field, the exchange bias modifies the velocity of current-driven domain wall motion by a factor of ten. We also find that the exchange bias remains stable under external fields up to 15 kOe and ns-long current pulses with current density up to 3.5x10^12 A/m^2. Our results demonstrate versatile control of the domain wall motion by exchange bias, which is relevant to achieve field-free switching of the magnetization in perpendicular systems and current-driven manipulation of domain walls velocity in spintronic devices.

Current-induced spin-orbit torques provide a versatile tool for switching magnetic devices. In perpendicular magnets, the dampinglike component of the torque is the main driver of magnetization reversal. The degree to which the fieldlike torque assists the switching is a matter of debate. Here we study the switching of magnetic tunnel junctions with a CoFeB free layer and either W or Ta underlayers, which have a ratio of fieldlike to dampinglike torque of 0.3 and 1, respectively. We show that the fieldlike torque can either assist or hinder the switching of CoFeB when the static in-plane magnetic field required to define the polarity of spin-orbit torque switching has a component transverse to the current. In particular, the non-collinear alignment of the field and current can be exploited to increase the switching efficiency and reliability compared to the standard collinear alignment. By probing individual switching events in real-time, we also show that the combination of transverse magnetic field and fieldlike torque can accelerate or decelerate the reversal onset. We validate our observations using micromagnetic simulations and extrapolate the results to materials with different torque ratios. Finally, we propose device geometries that leverage the fieldlike torque for density increase in memory applications and synaptic weight generation.

We measured the spin-orbit torques (SOTs), current-induced switching, and domain wall (DW) motion in synthetic ferrimagnets consisting of Co/Tb layers with differing stacking order grown on a Pt underlayer. We find that the SOTs, magnetic anisotropy, compensation temperature and SOT-induced switching are highly sensitive to the stacking order of Co and Tb and to the element in contact with Pt. Our study further shows that Tb is an efficient SOT generator when in contact with Co, such that its position in the stack can be adjusted to generate torques additive to those generated by Pt. With optimal stacking and layer thickness, the dampinglike SOT efficiency reaches up to 0.3, which is more than twice that expected from the Pt/Co bilayer. Moreover, the magnetization can be easily switched by the injection of pulses with current density of about 0.5-2*107A/cm2 despite the extremely high perpendicular magnetic anisotropy barrier (up to 7.8 T). Efficient switching is due to the combination of large SOTs and low saturation magnetization owing to the ferrimagnetic character of the multilayers. We observed current-driven DW motion in the absence of any external field, which is indicative of homochiral N\'eel-type DWs stabilized by the interfacial Dzyaloshinkii-Moriya interaction. These results show that the stacking order in transition metal/rare-earth synthetic ferrimagnets plays a major role in determining the magnetotransport properties relevant for spintronic applications.

Electron paramagnetic resonance (EPR) can provide unique insight into the chemical structure and magnetic properties of dopants in oxide and semiconducting materials that are of interest for applications in electronics, catalysis, and quantum sensing. Here, we demonstrate that EPR in combination with scanning tunneling microscopy (STM) allows for probing the bonding and charge state of alkali metal atoms on an ultrathin magnesium oxide layer on a Ag substrate. We observe a magnetic moment of $1\mu_\mathrm{B}$ for Li$_2$, LiNa, and Na$_2$ dimers corresponding to spin radicals with a charge state of $+1e$. Single alkali atoms have the same charge state and no magnetic moment. The ionization of the adsorbates is attributed to charge transfer through the oxide to the metal substrate. Our work highlights the potential of EPR-STM to provide insight into dopant atoms that are relevant for the control of the electrical properties of surfaces and nanodevices., Comment: 18 pages, 5 figures

The manipulation of antiferromagnetic order by means of spin-orbit torques opens unprecedented opportunities to exploit the dynamics of antiferromagnets in spintronic devices. In this work, we investigate the current-induced switching of the magnetic octupole vector in the Weyl antiferromagnet Mn$_3$Sn as a function of pulse shape, field, temperature, and time. We find that the switching behavior can be either bistable or tristable depending on the temporal structure of the current pulses. Time-resolved Hall effect measurements reveal that Mn$_3$Sn switching proceeds via a two-step demagnetization-remagnetization process caused by self-heating over a timescale of tens of ns followed by cooling in the presence of spin-orbit torques. Our results shed light on the switching dynamics of Mn$_3$Sn and prove the existence of extrinsic limits on its switching speed., Comment: Rectified wrong order of MS and Supplement

Many creative generative design spaces contain multiple regions with individuals of high aesthetic value. Yet traditional evolutionary computing methods typically focus on optimisation, searching for the fittest individual in a population. In this paper we apply quality-diversity search methods to explore a creative generative system (an agent-based line drawing model). We perform a random sampling of genotype space and use individual artist-assigned evaluations of aesthetic quality to formulate a computable fitness measure specific to the artist and this system. To compute diversity we use a convolutional neural network to discriminate features that are dimensionally reduced into two dimensions. We show that the quality-diversity search is able to find multiple phenotypes of high aesthetic value. These phenotypes show greater diversity and quality than those the artist was able to find using manual search methods., Comment: Preprint of paper accepted for EvoMUSART 2022 Conference

FPGA-based accelerators are becoming more popular for deep neural network due to the ability to scale performance with increasing degree of specialization with dataflow architectures or custom data types. To reduce the barrier for software engineers and data scientists to adopt FPGAs, C++- and OpenCL-based design entries with high-level synthesis (HLS) have been introduced. They provide higher abstraction compared to register-transfer level (RTL)-based design. HLS offers faster development time, better maintainability and more flexibility in code exploration, when evaluating options for multi-dimension tensors, convolutional layers or parallelism. Thus, HLS has been adopted by DNN accelerator generation frameworks such as FINN and hls4ml. In this paper, we present an alternative backend RTL library for FINN. We investigate and evaluate, across a spectrum of design dimensions, an RTL-based implementation versus the original HLS variant. We show that for smaller design parameters, RTL produces significantly smaller circuits. For larger circuits, however, the look-up table (LUT) count of RTL-based design is slightly higher, up to around $15\%$. On the other hand, HLS consistently requires more flip-flops (FFs) (orders-of-magnitude increase) and block RAMs (BRAMs) ($2\times$ more). This also impacts the critical path delay, with RTL producing significantly faster circuits, up to $80\%$. Furthermore, RTL also benefits from at-least a $10\times$ reduction in synthesis time. Finally the results were practically validated using a real-world use case of a multi-layer perceptron (MLP) network used in network intrusion detection. Overall, since HLS frameworks code-generate the hardware design, the benefits of the ease in the design entry is less important as compared to synthesis time reduction togther with resource benefits, this might make the RTL abstraction an attractive alternative., Comment: 22 pages, 7 tables, 16 figures

The possibility of electrically exciting and detecting magnon currents in magnetic insulators has opened exciting perspectives for transporting spin information in electronic devices. However, the role of the magnetic field and the nonlocal thermal gradients on the magnon transport remain unclear. Here, by performing nonlocal harmonic voltage measurements, we investigate magnon transport in perpendicularly magnetized ultrathin Tm3Fe5O12 (TmIG) films coupled to Pt electrodes. We show that the first harmonic nonlocal voltage captures spin-driven magnon transport in TmIG, as expected, and the second harmonic is dominated by thermoelectric voltages driven by current-induced thermal gradients at the detector. The magnon diffusion length in TmIG is found to be on the order of 0.3 {\mu}m at 0.5 T and gradually decays to 0.2 {\mu}m at 0.8 T, which we attribute to the suppression of the magnon relaxation time due to the increase of the Gilbert damping with field. By performing current, magnetic field, and distance dependent nonlocal and local measurements we demonstrate that the second harmonic nonlocal voltage exhibits five thermoelectric contributions, which originate from the nonlocal spin Seebeck effect and the ordinary, planar, spin, and anomalous Nernst effects. Our work provides a guide on how to disentangle magnon signals from diverse thermoelectric voltages of spin and magnetic origin in nonlocal magnon devices, and establish the scaling laws of the thermoelectric voltages in metal/insulator bilayers., Comment: 27 pages, 11 figures

Scanning magnetometry with nitrogen-vacancy (NV) centers in diamond has emerged as a powerful microscopy for studying weak stray field patterns with nanometer resolution. Due to the internal crystal anisotropy of the spin defect, however, external bias fields -- critical for the study of magnetic materials -- must be applied along specific spatial directions. In particular, the most common diamond probes made from {100}-cut diamond only support fields at an angle of $\theta = 55^\circ$ from the surface normal. In this paper, we report fabrication of scanning diamond probes from {110}-cut diamond where the spin anisotropy axis lies in the scan plane ($\theta=90^\circ$). We show that these probes retain their sensitivity in large in-plane fields and demonstrate scanning magnetometry of the domain pattern of Co-NiO films in applied fields up to 40 mT. Our work extends scanning NV magnetometry to the important class of materials that require large in-plane fields., Comment: 6 pages, 3 figures

The frequency band 1-15 GHz provides exciting prospects for resonant axion haloscopes as indicated by cosmological and astrophysical arguments. Among the challenges currently addressed to reach the required sensitivity, the development of high quality factor cavities that tolerate multi-Tesla fields plays a central role. We report a 3D resonator based on a right circular copper cavity with hollow cylinders that confine higher order modes around the cylinder axis. Its effective volume at 10.3\,GHz is $3.4 \cdot 10^{-2}$ liters, and under an 8\,T-field we measured an internal quality factor of more than 9 millions. These parameters demonstrate the potential of this unique resonator to probe galactic dark matter axion at remarkable scan rates of 15\,MHz/day when the cavity is readout by a quantum-limited receiver.

In this paper we examine the concept of complexity as it applies to generative and evolutionary art and design. Complexity has many different, discipline specific definitions, such as complexity in physical systems (entropy), algorithmic measures of information complexity and the field of "complex systems". We apply a series of different complexity measures to three different evolutionary art datasets and look at the correlations between complexity and individual aesthetic judgement by the artist (in the case of two datasets) or the physically measured complexity of generative 3D forms. Our results show that the degree of correlation is different for each set and measure, indicating that there is no overall "better" measure. However, specific measures do perform well on individual datasets, indicating that careful choice can increase the value of using such measures. We then assess the value of complexity measures for the audience by undertaking a large-scale survey on the perception of complexity and aesthetics. We conclude by discussing the value of direct measures in generative and evolutionary art, reinforcing recent findings from neuroimaging and psychology which suggest human aesthetic judgement is informed by many extrinsic factors beyond the measurable properties of the object being judged., Comment: Preprint of paper submitted to the Journal Genetic Programming and Evolvable Machines. arXiv admin note: substantial text overlap with arXiv:2102.02332

The establishment of chiral coupling in thin magnetic films with inhomogeneous anisotropy has led to the development of artificial systems of fundamental and technological interest. The chiral coupling itself is enabled by the Dzyaloshinskii-Moriya interaction (DMI) enforced by the patterned noncollinear magnetization. Here, we create a domain wall track with out-of-plane magnetization coupled on each side to a narrow parallel strip with in-plane magnetization. With this we show that the chiral torques emerging from the DMI at the boundary between the regions of noncollinear magnetization in a single magnetic layer can be used to bias the domain wall velocity. To tune the chiral torques, the design of the magnetic racetracks can be modified by varying the width of the tracks or the width of the transition region between noncollinear magnetizations, reaching effective chiral magnetic fields of up to 7.8 mT. Furthermore, we show how the magnitude of the chiral torques can be estimated by measuring asymmetric domain wall velocities, and demonstrate spontaneous domain wall motion propelled by intrinsic torques even in the absence of any external driving force.

We report on the occurrence of strong interlayer Dzyaloshinskii-Moriya interaction (DMI) between an in-plane magnetized Co layer and a perpendicularly magnetized TbFe layer through a Pt spacer. The DMI causes a chiral coupling that favors one-handed orthogonal magnetic configurations of Co and TbFe, which we reveal through Hall effect and magnetoresistance measurements. The DMI coupling mediated by Pt causes effective magnetic fields on either layer of up to 10-15 mT, which decrease monotonously with increasing Pt thickness. Ru, Ta, and Ti spacers mediate a significantly smaller coupling compared to Pt, highlighting the essential role of Pt in inducing the interlayer DMI. These results are relevant to understand and maximize the interlayer coupling induced by the DMI as well as to design spintronic devices with chiral spin textures.

Electron paramagnetic resonance (EPR) performed with a scanning tunneling microscope (STM) allows for probing the spin excitation of single atomic species with MHz energy resolution. One of the basic applications of conventional EPR is the precise determination of magnetic moments. However, in an STM, the local magnetic fields of the spin-polarized tip can introduce systematic errors in the measurement of the magnetic moments by EPR. We propose to solve this issue by finding tip-sample distances at which the EPR resonance shift caused by the magnetic field of the tip is minimized. To this end, we measure the dependence of the resonance field on the tip-sample distance at different radiofrequencies and identify specific distances for which the true magnetic moment is found. Additionally, we show that the tip's influence can be averaged out by using magnetically bistable tips, which provide a complementary method to accurately measure the magnetic moment of surface atoms using EPR-STM.

We consider the problem of detecting, in the visual sensing data stream of an autonomous mobile robot, semantic patterns that are unusual (i.e., anomalous) with respect to the robot's previous experience in similar environments. These anomalies might indicate unforeseen hazards and, in scenarios where failure is costly, can be used to trigger an avoidance behavior. We contribute three novel image-based datasets acquired in robot exploration scenarios, comprising a total of more than 200k labeled frames, spanning various types of anomalies. On these datasets, we study the performance of an anomaly detection approach based on autoencoders operating at different scales.

We consider the task of visually estimating the pose of a human from images acquired by a nearby nano-drone; in this context, we propose a data augmentation approach based on synthetic background substitution to learn a lightweight CNN model from a small real-world training set. Experimental results on data from two different labs proves that the approach improves generalization to unseen environments.

We propose the use of probabilistic programming techniques to tackle the malicious user identification problem in a recommendation algorithm. Probabilistic programming provides numerous advantages over other techniques, including but not limited to providing a disentangled representation of how malicious users acted under a structured model, as well as allowing for the quantification of damage caused by malicious users. We show experiments in malicious user identification using a model of regular and malicious users interacting with a simple recommendation algorithm, and provide a novel simulation-based measure for quantifying the effects of a user or group of users on its dynamics., Comment: Short version, appeared at ICML workshop on Socially Responsible Machine Learning 2021