Power and area efficient stochastic artificial neural networks using spin–orbit torque-based true random number generator

Abstract

Hardware implementations of Artificial Neural Networks (ANNs) using conventional binary arithmetic units are computationally expensive and energy-intensive together with large area footprints. Stochastic computing (SC) is an unconventional computing paradigm that operates on stochastic bit streams. It can offer low-power and area-efficient hardware implementations and has shown promising results when applied to ANN hardware circuits. SC relies on stochastic number generators (SNGs) to map input binary numbers to stochastic bit streams. The SNGs are conventionally implemented using random number generators (RNGs) and comparators. Linear feedback shifted registers (LFSRs) are typically used as the RNGs, which need far more area and power than the SC core, counteracting the latter's main advantages. To mitigate this problem, in this Letter, RNGs employing Spin–Orbit Torque (SOT)-induced stochastic switching of perpendicularly magnetized Ta/CoFeB/MgO nanodevices have been proposed. Furthermore, the SOT true random number generator (TRNG) is integrated with the simple CMOS stochastic computing circuits to perform a stochastic artificial neural network. To further optimize power and area efficiency, a fully parallel architecture and TRNG-sharing scheme are presented. The proposed stochastic ANN using the SOT-based TRNG obtains a negligible inference accuracy loss, compared with the binary version, and achieves 9× and 25× improvement in terms of area and power, respectively, compared with the ANN using LFSRs.

Application of Resistive Random Access Memory in Hardware Security: A Review

“…current pulse's amplitude and duration is required in the former while the latter is affected by the temperature variations. The stochasticity in the switching of SOT-MRAM is also currently being explored for TRNG, [180][181][182] and similar to STT-MRAM, the robustness of such devices is still a downside. By applying a voltage pulse for which the probability of switching is 50%, the TRNG was demonstrated with FeFET.…”

Reduction of Operating Current by Harnessing the Field‐ and Damping‐Like Torque Ratios in Nonmagnet–Ferromagnet Heterojunctions

“…metal-oxide-semiconductor (CMOS)-based RNGs that require postprocessing steps to ensure randomness. [19][20][21][22][23][24] Prior SOT studies ignore the influence of FLT on the switching process. Most previous studies have not analyzed the FLT impact, focusing solely on analyzing DLT during switching operations.…”

“…As a result, active exploration of SOT‐based TRNGs, which leverage intrinsic stochastic properties, is underway to replace the complementary metal–oxide–semiconductor (CMOS)‐based RNGs that require postprocessing steps to ensure randomness. [ 19–24 ]…”

Switching plasticity in compensated ferrimagnetic multilayers for neuromorphic computing

“…[5,6] Although various artificial neural networks algorithms have been developed in recent years, the lack of dedicated hardware still limits the application of neuromorphic computing. [7] Recently, spintronic devices have exhibited great potential in neuromorphic computing, [8][9][10][11][12][13][14][15][16][17][18][19] since they can simulate the functions of neurons and synapses, such as nonlinearity, [9] stochasticity [8,20] and nonvolatility. [1,21] Moreover, the fast dynamics [22,23] and virtually unlimited endurance make them stand out from other competitors including phase-change, [24,25] floating gated [26] and resistive memory [6] devices.…”