Stochastic Spiking Neural Networks Enabled by Magnetic Tunnel Junctions: From Nontelegraphic to Telegraphic Switching Regimes

Abstract

Stochastic spiking neural networks based on nanoelectronic spin devices can be a possible pathway to achieving "brainlike" compact and energy-effcient cognitive intelligence. The computational model attempt to exploit the intrinsic device stochasticity of nanoelectronic synaptic or neural components to perform learning or inference. However, there has been limited analysis on the scaling effect of stochastic spin devices and its impact on the operation of such stochastic networks at the system level. This work attempts to explore the design space and analyze the performance of nanomagnet-based stochastic neuromorphic computing architectures for magnets with different barrier heights. We illustrate how the underlying network architecture must be modified to account for the random telegraphic switching behavior displayed by magnets with low barrier heights as they are scaled into the superparamagnetic regime. We perform a device-to-system-level analysis on a deep neural-network architecture for a digit-recognition problem on the MNIST data set.

arXiv comment12 pages, 14 Figures, 1 Table

References in corpus (9)

• Spin torque switching with the giant spin Hall effect of tantalum
• Training and Operation of an Integrated Neuromorphic Network Based on Metal-Oxide Memristors
• Spin transfer torque devices utilizing the giant spin Hall effect of tungsten
• Intrinsic optimization using stochastic nanomagnets
• Boltzmann Test of Slonczewski's Theory of Spin Transfer Torque
• Biologically Inspired Spiking Neurons : Piecewise Linear Models and Digital Implementation
• Probabilistic Deep Spiking Neural Systems Enabled by Magnetic Tunnel Junction
• Voltage and Energy-Delay Performance of Giant Spin Hall Effect Switching for Magnetic Memory and Logic
• Hybrid Spintronic-CMOS Spiking Neural Network With On-Chip Learning: Devices, Circuits and Systems

Cited by in corpus (12)

• p-Bits for Probabilistic Spin Logic
• Skyrmions in magnetic tunnel junctions
• Low Barrier Magnet Design for Efficient Hardware Binary Stochastic Neurons
• Autonomous Probabilistic Coprocessing with Petaflips per Second
• Composable Probabilistic Inference Networks Using MRAM-based Stochastic Neurons
• Double Free-Layer Magnetic Tunnel Junctions for Probabilistic Bits
• Stochastic Magnetoelectric Neuron for Temporal Information Encoding
• Overcoming device unreliability with continuous learning in a population coding based computing system
• An All-Memristor Deep Spiking Neural Computing System: A Step Towards Realizing the Low Power,Stochastic Brain
• Measurement-driven neural-network training for integrated magnetic tunnel junction arrays
• Ground-State Probabilistic Logic with the Simplest Binary Energy Landscape for Probabilistic Computing
• Kramers' escape problem for white noise driven switching in ferroelectrics