Hybrid Piezoelectric-Magnetic Neurons: A Proposal for Energy-Efficient Machine Learning

TL;DR

This paper introduces a hybrid spintronic neuron combining piezoelectric and magnetic components, demonstrating significant energy efficiency improvements and smaller size compared to traditional CMOS neurons, making it promising for neural network applications.

Contribution

It proposes a novel hybrid piezoelectric-magnetic neuron structure and simulates its energy-efficient operation, showing advantages over existing spintronic and CMOS neurons.

Findings

70% energy dissipation reduction compared to other spintronic neurons

Smaller footprint and lower energy consumption than CMOS neurons

Potential for integration into artificial neural networks

Abstract

This paper proposes a spintronic neuron structure composed of a heterostructure of magnets and a piezoelectric with a magnetic tunnel junction (MTJ). The operation of the device is simulated using SPICE models. Simulation results illustrate that the energy dissipation of the proposed neuron compared to that of other spintronic neurons exhibits 70% improvement. Compared to CMOS neurons, the proposed neuron occupies a smaller footprint area and operates using less energy. Owing to its versatility and low-energy operation, the proposed neuron is a promising candidate to be adopted in artificial neural network (ANN) systems.