Search NASA⌕ Search

Engineering topics

Liu, Samuel

Publications and source records attributed to Liu, Samuel.

17 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Magnetic tunnel junction random number generators applied to dynamically tuned probability trees driven by spin orbit torque

Abstract Perpendicular magnetic tunnel junction (pMTJ)-based true-random number generators (RNGs) can consume orders of magnitude less energy per bit than CMOS pseudo-RNGs. Here, we numerically investigate with a macrospin Landau–Lifshitz-Gilbert equation solver the use of pMTJs driven by spin–orbit torque to directly sample numbers from arbitrary probability distributions with the help of a tunable probability tree. The tree operates by dynamically biasing sequences of pMTJ relaxation events, called ‘coinflips’, via an additional applied spin-transfer-torque current. Specifically, using a single, ideal pMTJ device we successfully draw integer samples on the interval [0, 255] from an exponential distribution based on p -value distribution analysis. In order to investigate device-to-device variations, the thermal stability of the pMTJs are varied based on manufactured device data. It is found that while repeatedly using a varied device inhibits ability to recover the probability distribution, the device variations average out when considering the entire set of devices as a ‘bucket’ to agnostically draw random numbers from. Further, it is noted that the device variations most significantly impact the highest level of the probability tree, with diminishing errors at lower levels. The devices are then used to draw both uniformly and exponentially distributed numbers for the Monte Carlo computation of a problem from particle transport, showing excellent data fit with the analytical solution. Finally, the devices are benchmarked against CMOS and memristor RNGs, showing faster bit generation and significantly lower energy use.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2022

Shape‐Dependent Multi‐Weight Magnetic Artificial Synapses for Neuromorphic Computing

Abstract In neuromorphic computing, artificial synapses provide a multi‐weight (MW) conductance state that is set based on inputs from neurons, analogous to the brain. Herein, artificial synapses based on magnetic materials that use a magnetic tunnel junction (MTJ) and a magnetic domain wall (DW) are explored. By fabricating lithographic notches in a DW track underneath a single MTJ, 3–5 stable resistance states that can be repeatably controlled electrically using spin‐orbit torque are achieved. The effect of geometry on the synapse behavior is explored, showing that a trapezoidal device has asymmetric weight updates with high controllability, while a rectangular device has higher stochasticity, but with stable resistance levels. The device data is input into neuromorphic computing simulators to show the usefulness of application‐specific synaptic functions. Implementing an artificial neural network (NN) applied to streamed Fashion‐MNIST data, the trapezoidal magnetic synapse can be used as a metaplastic function for efficient online learning. Implementing a convolutional NN for CIFAR‐100 image recognition, the rectangular magnetic synapse achieves near‐ideal inference accuracy, due to the stability of its resistance levels. This work shows MW magnetic synapses are a feasible technology for neuromorphic computing and provides design guidelines for emerging artificial synapse technologies.

Leonard, Thomas↗
DOE OSTI2022

Probabilistic Nanomagnetic Memories for Uncertain and Robust Machine Learning

This project evaluated the use of emerging spintronic memory devices for robust and efficient variational inference schemes. Variational inference (VI) schemes, which constrain the distribution for each weight to be a Gaussian distribution with a mean and standard deviation, are a tractable method for calculating posterior distributions of weights in a Bayesian neural network such that this neural network can also be trained using the powerful backpropagation algorithm. Our project focuses on domain-wall magnetic tunnel junctions (DW-MTJs), a powerful multi-functional spintronic synapse design that can achieve low power switching while also opening the pathway towards repeatable, analog operation using fabricated notches. Our initial efforts to employ DW-MTJs as an all-in-one stochastic synapse with both a mean and standard deviation didn’t end up meeting the quality metrics for hardware-friendly VI. In the future, new device stacks and methods for expressive anisotropy modification may make this idea still possible. However, as a fall back that immediately satisfies our requirements, we invented and detailed how the combination of a DW-MTJ synapse encoding the mean and a probabilistic Bayes-MTJ device, programmed via a ferroelectric or ionically modifiable layer, can robustly and expressively implement VI. This design includes a physics-informed small circuit model, that was scaled up to perform and demonstrate rigorous uncertainty quantification applications, up to and including small convolutional networks on a grayscale image classification task, and larger (Residual) networks implementing multi-channel image classification. Lastly, as these results and ideas all depend upon the idea of an inference application where weights (spintronic memory states) remain non-volatile, the retention of these synapses for the notched case was further interrogated. These investigations revealed and emphasized the importance of both notch geometry and anisotropy modification in order to further enhance the endurance of written spintronic states. In the near future, these results will be mapped to effective predictions for room temperature and elevated operation DW-MTJ memory retention, and experimentally verified when devices become available.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2022

Metaplastic and energy-efficient biocompatible graphene artificial synaptic transistors for enhanced accuracy neuromorphic computing

CMOS-based computing systems that employ the von Neumann architecture are relatively limited when it comes to parallel data storage and processing. In contrast, the human brain is a living computational signal processing unit that operates with extreme parallelism and energy efficiency. Although numerous neuromorphic electronic devices have emerged in the last decade, most of them are rigid or contain materials that are toxic to biological systems. In this work, we report on biocompatible bilayer graphene-based artificial synaptic transistors (BLAST) capable of mimicking synaptic behavior. The BLAST devices leverage a dry ion-selective membrane, enabling long-term potentiation, with ~50 aJ/µm 2 switching energy efficiency, at least an order of magnitude lower than previous reports on two-dimensional material-based artificial synapses. The devices show unique metaplasticity, a useful feature for generalizable deep neural networks, and we demonstrate that metaplastic BLASTs outperform ideal linear synapses in classic image classification tasks. With switching energy well below the 1 fJ energy estimated per biological synapse, the proposed devices are powerful candidates for bio-interfaced online learning, bridging the gap between artificial and biological neural networks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

A domain wall-magnetic tunnel junction artificial synapse with notched geometry for accurate and efficient training of deep neural networks

Inspired by the parallelism and efficiency of the brain, several candidates for artificial synapse devices have been developed for neuromorphic computing, yet a nonlinear and asymmetric synaptic response curve precludes their use for backpropagation, the foundation of modern supervised learning. Spintronic devices—which benefit from high endurance, low power consumption, low latency, and CMOS compatibility—are a promising technology for memory, and domain-wall magnetic tunnel junction (DW-MTJ) devices have been shown to implement synaptic functions such as long-term potentiation and spike-timing dependent plasticity. In this work, we propose a notched DW-MTJ synapse as a candidate for supervised learning. Using micromagnetic simulations at room temperature, we show that notched synapses ensure the non-volatility of the synaptic weight and allow for highly linear, symmetric, and reproducible weight updates using either spin transfer torque (STT) or spin–orbit torque (SOT) mechanisms of DW propagation. We use lookup tables constructed from micromagnetics simulations to model the training of neural networks built with DW-MTJ synapses on both the MNIST and Fashion-MNIST image classification tasks. Accounting for thermal noise and realistic process variations, the DW-MTJ devices achieve classification accuracy close to ideal floating-point updates using both STT and SOT devices at room temperature and at 400 K. Our work establishes the basis for a magnetic artificial synapse that can eventually lead to hardware neural networks with fully spintronic matrix operations implementing machine learning.

42 ENGINEERING↗
DOE OSTI2021

Controllable Reset Behavior in Domain Wall–Magnetic Tunnel Junction Artificial Neurons for Task-Adaptable Computation

Neuromorphic computing with spintronic devices has been of interest due to the limitations of CMOS-driven von Neumann computing. Domain wall–magnetic tunnel junction (DW-MTJ) devices have been shown to be able to intrinsically capture biological neuron behavior. Edgy-relaxed behavior, where a frequently firing neuron experiences a lower action potential threshold, may provide additional artificial neuronal functionality when executing repeated tasks. In this letter, we demonstrate that this behavior can be implemented in DW-MTJ artificial neurons via three alternative mechanisms: shape anisotropy, magnetic field, and current-driven soft reset. Using micromagnetics and analytical device modeling to classify the Optdigits handwritten digit dataset, we show that edgy-relaxed behavior improves both classification accuracy and classification rate for ordered datasets while sacrificing little to no accuracy for a randomized dataset. This letter establishes methods by which artificial spintronic neurons can be flexibly adapted to datasets.

42 ENGINEERING↗
DOE OSTI2013

Ordered vs. disordered perovskites; structural studies of Fe-doped SrIrO{sub 3} and SrRuO{sub 3}

The structures of the two Fe containing perovskites Sr{sub 2}IrFeO{sub 6} and SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} have been established using a combination of synchrotron and neutron diffraction methods. Sr{sub 2}IrFeO{sub 6} and SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} are shown to be monoclinic I2/m and tetragonal I4/mcm respectively The former exhibits a rock-salt like ordering of the Fe and Ir cations and displays a sequence of phase transitions associated with the loss of the octahedral tilts upon heating; 12/m→I4/m→Fm3{sup ¯}m. The Fe and Ru cations are disordered in SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} and this shows a single structural phase transition upon heating due to the loss of the in-phase tilts, viz. I4/mcm→Pm3{sup ¯}m. In both cases XANES measurements show partial oxidation of the Fe{sup 3+} to Fe{sup 4+}. The difference in the structures between the two is remarkable given the similar size of Ir{sup 5+} and Ru{sup 5+}, and this is reflected in their magnetic properties. - Graphical abstract: Sr{sub 2}IrFeO{sub 6} and SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} are shown, using a combination of synchrotron and neutron diffraction, to be monoclinic I2/m with cation ordering and tetragonal I4/mcm with disordered Fe and Ir, respectively. Both undergo phase transitions upon heating due to the loss of the octahedral tilts. Display Omitted - Highlights: • Sr{sub 2}IrFeO{sub 6} shown to be monoclinic and shows the transitions upon heating I2/m→I4/m→ Fm3{sup ¯}m. • SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} is tetragonal and shows a single I4/mcm→Pm3m transition upon heating. • The Fe and Ru cations are disordered in SrRu{sub 0.5}Fe{sub 0.5}O{sub 3} but ordered in Sr{sub 2}FeIrO{sub 6}. • XANES measurements show partial oxidation of the Fe{sup 3+} to Fe{sup 4+}.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2013

Designing new n=2 Sillen–Aurivillius phases by lattice-matched substitutions in the halide and [Bi{sub 2}O{sub 2}]{sup 2+} layers

A new n=2 Sillen–Aurivillius compound Bi{sub 3}Sr{sub 2}Nb{sub 2}O{sub 11}Br has been synthesised based on Bi{sub 3}Pb{sub 2}Nb{sub 2}O{sub 11}Cl by simultaneously replacing Pb{sup 2+} with Sr{sup 2+} and Cl{sup −} with Br{sup −}. Rietveld refinements against X-ray and neutron powder diffraction data revealed a significant relative compression in the stacking axis (c-axis) of the new compound. Sr{sup 2+} doping reduces the impact of the stereochemically active 6s{sup 2} lone pair found on Pb{sup 2+} and Bi{sup 3+}, resulting in a contraction of the c-axis by 1.22% and an expansion of the ab plane by 0.25%. This improves the inter-layer compatibility with the larger halide Br{sup −}. Analysis of X-ray absorption near-edge spectroscopy data show that the ferroelectric distortion of the B-site cation is less apparent in Bi{sub 3}Sr{sub 2}Nb{sub 2}O{sub 11}Br compared to Bi{sub 3}Pb{sub 2}Nb{sub 2}O{sub 11}Cl, and variable-temperature neutron diffraction data show no evidence for a ferroelectric distortion. - Graphical abstract: The structural expansion and contractions in the stacking axis of co-doped Sillen–Aurivillius phases. Display Omitted - Highlights: • A new Sillen–Aurivillius phase Bi{sub 3}Sr{sub 2}Nb{sub 2}O{sub 11}Br is synthesised based on Bi{sub 3}Pb{sub 2}Nb{sub 2}O{sub 11}Cl. • The stacking axis contracts although local expansion accommodates the larger halide Br. • Majority of the contraction is observed in the [Bi{sub 2}O{sub 2}]{sup 2+} layers where we replaced Pb{sup 2+} for Sr{sup 2+}. • Removing the Pb{sup 2+} lowers the lone pair effect and improves inter-layer compatibility. • XANES and NPD suggest that the ferroelectric properties are minimised or suppressed.

36 MATERIALS SCIENCE↗