Reliability of Neural Networks Based on Spintronic Neurons

Spintronic technology is emerging as a direction for the hardware implementation of neurons and synapses of neuromorphic architectures. In particular, a single spintronic device can be used to implement the nonlinear activation function of neurons. Here, we present how to implement spintronic neurons with sigmoidal and rectified linear unit (ReLU)-like activation functions. We then perform a numerical experiment showing the reliability of neural networks made by spintronic neurons, all having different activation functions to emulate device-to-device variations in a possible hardware implementation of the network. Therefore, we consider a vanilla'' neural network implemented to recognize the categories of the Mixed National Institute of Standards and Technology database, and we show an average accuracy of 98.87% in the test dataset, which is very close to 98.89% as obtained for the ideal case (all neurons have the same sigmoid activation function). Similar results are obtained with neurons having a ReLU-like activation function.

Automated summary

The study demonstrates how to implement spintronic neurons with different activation functions and conducts a numerical experiment to validate the reliability of neural networks, showing an average accuracy of 98.87% in recognizing the Mixed National Institute of Standards and Technology database in the test dataset, which is very close to the ideal case where all neurons have the same sigmoid activation function.