Related papers: Reconfigurable classifier based on spin torque dri…
Solid state spin qubits are promising candidates for quantum information processing, but controlled interactions and entanglement in large, multi-qubit systems are currently difficult to achieve. We describe a method for programmable…
This chapter presents a review on spin transfer torque in magnetic tunnel junctions. In the first part, we propose an overview of experimental and theoretical studies addressing current-induced magnetization excitations in magnetic tunnel…
Rapid random telegraph noise (RTN) in magnetic tunnel junctions (MTJs) is an important figure of merit for probabilistic computing applications. However, the interactions between the macrospin and spin waves with finite wave numbers reduce…
Control of magnetism without using magnetic fields enables large-scale integration of spintronic devices for memory, computation and communication in the beyond-CMOS era. Mechanisms including spin torque transfer, spin Hall effect, and…
Neural machine translation (NMT) heavily relies on word-level modelling to learn semantic representations of input sentences. However, for languages without natural word delimiters (e.g., Chinese) where input sentences have to be tokenized…
Neuromorphic computing, inspired by the brain's parallel and energy-efficient processing, offers a transformative approach to artificial intelligence. In this study, we fabricated optimized spin-transfer torque nano-oscillators (STNOs) and…
We present artificial neural network design using spin devices that achieves ultra low voltage operation, low power consumption, high speed, and high integration density. We employ spin torque switched nano-magnets for modelling neuron and…
We present a simple and fast method to simulate spin-torque driven magnetisation dynamics in nano-pillar spin-valve structures. The approach is based on the coupling between a spin transport code based on random matrix theory and a…
We address a learning-to-normalize problem by proposing Switchable Normalization (SN), which learns to select different normalizers for different normalization layers of a deep neural network. SN employs three distinct scopes to compute…
Magnetic skyrmions, as scalable and non-volatile spin textures, can dynamically interact with fields and currents, making them promising for unconventional computing. This paper presents a neuromorphic device based on skyrmion manipulation…
Neural machine translation (NMT) models are typically trained with fixed-size input and output vocabularies, which creates an important bottleneck on their accuracy and generalization capability. As a solution, various studies proposed…
Dedicated hardware implementations of spiking neural networks that combine the advantages of mixed-signal neuromorphic circuits with those of emerging memory technologies have the potential of enabling ultra-low power pervasive sensory…
Thermodynamic-driven filament formation in redox-based resistive memory and the impact of thermal fluctuations on switching probability of emerging magnetic switches are probabilistic phenomena in nature, and thus, processes of binary…
Neural interfaces capable of multi-site electrical recording, on-site signal classification, and closed-loop therapy are critical for the diagnosis and treatment of neurological disorders. However, deploying machine learning algorithms on…
In this paper, a spintronic neuromorphic reconfigurable Array (SNRA) is developed to fuse together power-efficient probabilistic and in-field programmable deterministic computing during both training and evaluation phases of restricted…
A three-dimensional self-consistent spin transport model is developed, which includes both tunnelling transport, as well as metallic transport. Using the spin accumulation computed either side of a tunnel barrier, spin torques are obtained,…
We present a computationally efficient strategy that allows to simulate magnetization switching driven by spin-transfer torque in magnetic tunnel junctions within a micromagnetic model coupled with a matrix-based non-equilibrium Green's…
In this work we propose a new supervised learning method for temporally-encoded multilayer spiking networks to perform classification. The method employs a reinforcement signal that mimics backpropagation but is far less computationally…
Most modern neural machine translation (NMT) systems rely on presegmented inputs. Segmentation granularity importantly determines the input and output sequence lengths, hence the modeling depth, and source and target vocabularies, which in…
Superparamagnetic tunnel junctions (SMTJs) have emerged as a competitive, realistic nanotechnology to support novel forms of stochastic computation in CMOS-compatible platforms. One of their applications is to generate random bitstreams…