A Novel Compound Synapse Using Probabilistic Spin–Orbit-Torque Switching for MTJ-Based Deep Neural Networks

Abstract

Analog electronic non-volatile memories mimicking synaptic operations are being explored for the implementation of neuromorphic computing systems. Compound synapses consisting of ensembles of stochastic binary elements are alternatives to analog memory synapses to achieve multilevel memory operation. Among existing binary memory technologies, magnetic tunneling junction (MTJ) based Magnetic Random Access Memory (MRAM) technology has matured to the point of commercialization. More importantly for this work, stochasticity is natural to the MTJ switching physics e.g devices referred as p-bits which mimic binary stochastic neurons. In this article, we experimentally demonstrate a novel compound synapse that uses stochastic spin-orbit torque (SOT) switching of an ensemble of nano-magnets that are located on one shared spin Hall effect (SHE) material channel, i.e. tantalum. By using a properly chosen pulse scheme, we are able to demonstrate linear potentiation and depression in the synapse, as required for many neuromorphic architectures. In addition to this experimental effort, we also performed circuit simulations on an SOT-MRAM based 784×200×10 deep belief network (DBN) consisting of p-bit based neurons and compound synapses. MNIST pattern recognition was used to evaluate the system performance, and our findings indicate that a significant reduction in recognition error rates can be achieved when using our incremental pulse scheme rather than a non-linear potentiation and depression as obtained when employing identical pulses.

Neuromorphic Engineering: From Biological to Spike‐Based Hardware Nervous Systems

“…By adopting appropriate pulse scheme, linear potentiation and depression of the synaptic weights were successfully demonstrated, which effectively improved the recognition rates of the MNIST patterns during deep belief network (DBN) simulations. [ 122 ] Tuning the efficacy of the SOT by the in‐plane field component can broaden a binary ferromagnet device into an analog synapse with multistate (Figure 4g–i). [ 123 ] A current pulse was utilized to manipulate the orientation of the magnetic material to successfully demonstrate EPSP, IPSP, and STDP.…”

Gradient Descent on Multilevel Spin–Orbit Synapses with Tunable Variations

“…[ 4–7 ] Spin–obit torque (SOT) devices [ 8,9 ] with inherent nonvolatility and radiation resistance, subnanosecond switching dynamics, unlimited endurance, excellent stability, and verified complementary metal‐oxide‐semiconductor transistor (CMOS)‐compatibility can also be implemented as the multilevel current‐induced magnetization switching, which can be used to emulate the synaptic plasticity. [ 10–18 ] Beyond the device level, artificial and spiking neural networks [ 19,20 ] consisting of spin–orbit synapses have been applied for various purposes such as associative memories, [ 21 ] deep belief learnings, [ 22 ] on‐the‐fly learnings, [ 23 ] etc. According to the studies on other emerging nonvolatile memories, especially the resistive switching random access memory (RRAM) and the phase change memory (PCM), [ 24,25 ] the variation and linearity of the device are crucial parameters for neuromorphic applications.…”

A survey of in-spin transfer torque MRAM computing

“…With respect to the circuit-level implementation of storage blocks, memory subsystems can perform the storage function as well as the associated arithmetic and computing units. IMC and NMC were investigated, and the majority of them underwent silicon verification in SRAM [4,[18][19][20][21][22] and several NVMs, including RRAM [23][24][25], STT-MRAM [26][27][28][29][30][31], spin-orbit torque (SOT), and MRAM [32][33][34].…”