Related papers: SOT-MRAM-Enabled Probabilistic Binary Neural Netwo…

Tensor networks (TNs) enable compact representations of large tensors through shared parameters. Their use in probabilistic modeling is particularly appealing, as probabilistic tensor networks (PTNs) allow for tractable computation of…

Binary Neural Networks (BNNs) are an extremely promising method to reduce deep neural networks' complexity and power consumption massively. Binarization techniques, however, suffer from ineligible performance degradation compared to their…

Reliable operation of perpendicular spin-transfer-torque magnetic random-access memory (p-STT-MRAM) requires control of magnetic alignment within the synthetic antiferromagnet (SAF) reference layer. At nanopillar dimensions, however,…

Preference-based Reinforcement Learning (PbRL) circumvents the need for reward engineering by harnessing human preferences as the reward signal. However, current PbRL methods excessively depend on high-quality feedback from domain experts,…

Neural-network (NN) inference is increasingly present on-board spacecraft to reduce downlink bandwidth and enable timely decision making. However, the power and reliability constraints of space missions limit the applicability of many…

To reduce the resource demand of neural network (NN) inference systems, it has been proposed to use approximate memory, in which the supply voltage and the timing parameters are tuned trading accuracy with energy consumption and…

Spin current and spin torque generation through the spin-orbit interactions in solids, of bulk or interfacial origin, is at the heart of spintronics research. The realization of spin-orbit torque (SOT) driven magnetic dynamics and switching…

Training and running deep neural networks (NNs) often demands a lot of computation and energy-intensive specialized hardware (e.g. GPU, TPU...). One way to reduce the computation and power cost is to use binary weight NNs, but these are…

Recently there has been increasing activity to build dedicated Ising Machines to accelerate the solution of combinatorial optimization problems by expressing these problems as a ground-state search of the Ising model. A common theme of such…

Stochasticity plays a significant role in the low-power operation of a biological neural network. In an artificial neural network (ANN), stochasticity also contributes to critical functions such as the uncertainty quantification (UQ) for…

Current-induced spin-orbit torques (SOTs) represent one of the most effective ways to manipulate the magnetization in spintronic devices. The orthogonal torque-magnetization geometry, the strong damping, and the large domain wall velocities…

Memristor based neural networks have great potentials in on-chip neuromorphic computing systems due to the fast computation and low-energy consumption. However, the imprecise properties of existing memristor devices generally result in…

Hardware-based spiking neural networks (SNNs) are regarded as promising candidates for the cognitive computing system due to low power consumption and highly parallel operation. In this work, we train the SNN in which the firing time…

Neuromorphic computing, inspired by the brain, promises extreme efficiency for certain classes of learning tasks, such as classification and pattern recognition. The performance and power consumption of neuromorphic computing depends…

Deep learning has emerged as an effective approach for creating modern software systems, with neural networks often surpassing hand-crafted systems. Unfortunately, neural networks are known to suffer from various safety and security issues.…

Deterministic magnetization switching using spin-orbit torque (SOT) has recently emerged as an efficient means to electrically control the magnetic state of ultrathin magnets. The SOT switching still lacks in oscillatory switching…

Pre-trained vision models have found widespread application across diverse domains. Prompt tuning-based methods have emerged as a parameter-efficient paradigm for adapting pre-trained vision models. While effective on standard benchmarks,…

Spintronic devices, such as non-volatile magnetic random access memories and logic devices, have attracted considerable attention as potential candidates for future high efficient data storage and computing technology. In a heavy metal or…

Convolutional Neural Networks (CNNs) demonstrate excellent performance in various applications but have high computational complexity. Quantization is applied to reduce the latency and storage cost of CNNs. Among the quantization methods,…

Spiking Neural Networks (SNNs) have recently emerged as the low-power alternative to Artificial Neural Networks (ANNs) owing to their asynchronous, sparse, and binary information processing. To improve the energy-efficiency and throughput,…