Spin switches for compact implementation of neuron and synapse

Abstract

Nanomagnets driven by spin currents provide a natural implementation for a neuron and a synapse: currents allow convenient summation of multiple inputs, while the magnet provides the threshold function. The objective of this paper is to explore the possibility of a hardware neural network (HNN) implementation using a spin switch (SS) as its basic building block. SS is a recently proposed device based on established technology with a transistor-like gain and input-output isolation. This allows neural networks to be constructed with purely passive interconnections without intervening clocks or amplifiers. The weights for the neural network are conveniently adjusted through analog voltages that can be stored in a non-volatile manner in an underlying CMOS layer using a floating gate low dropout voltage regulator. The operation of a multi-layer SS neural network designed for character recognition is demonstrated using a standard simulation model based on coupled Landau-Lifshitz-Gilbert (LLG) equations, one for each magnet in the network.

7 pages, 9 figures and 1 table

References in corpus (5)

• Study of Intrinsic Spin Hall Effect and Orbital Hall Effect in 4d- and 5d- Transition Metals
• Spin torque building blocks
• Memcomputing: a computing paradigm to store and process information on the same physical platform
• Non-volatile Spin Switch for Boolean and Non-Boolean Logic
• Spin switches for compact implementation of neuron and synapse

Cited by in corpus (13)

• Stochastic p-bits for Invertible Logic
• A Survey of Neuromorphic Computing and Neural Networks in Hardware
• Intrinsic optimization using stochastic nanomagnets
• Low Barrier Nanomagnets as p-bits for Spin Logic
• A building block for hardware belief networks
• Computing with spins and magnets
• Spin switches for compact implementation of neuron and synapse
• From materials to systems: a multiscale analysis of nanomagnetic switching
• Analog Signal Processing Using Stochastic Magnets
• Evaluating Spintronic Devices Using The Modular Approach
• Smart Detector Cell: A Scalable All-Spin Circuit for Low Power Non-Boolean Pattern Recognition
• The Roles of Kerr nonlinearity in a Bosonic Quantum Neural Network
• Hardware based Spatio-Temporal Neural Processing Backend for Imaging Sensors: Towards a Smart Camera