Controllable SiOx Nanorod Memristive Neuron for Probabilistic Bayesian Inference.

Modern artificial neural network technology using a deterministic computing framework is faced with a critical challenge in dealing with massive genuine data that are largely unstructured and ambiguous. This challenge demands the advances of an elementary physical device for tackling the uncertainties inherent to natural data, which can be a stepping stone for realizing probabilistic neural networks. Here, we designed and fabricated a SiOx nanorod memristive device by employing the glancing angle deposition (GLAD) technique, suggesting a controllable stochastic artificial neuron that can mimic the fundamental integrate-and-fire signaling and stochastic dynamics of a biological neuron. The SiOx nanorod structure provides the random distribution of multiple nanopores all across the active area, capable of forming a multitude of Si switching filaments at many SiOx nanorod edges after the electromigration process, leading to a stochastic switching event with a very high dynamic range (up to ∼5.15 × 1010 ) and low energy (up to ∼ 4.06 pJ) for firing. Different probabilistic activation (ProbAct) functions in a sigmoid form are implemented, showing its controllability with low variation by manufacturing and electrical programming schemes. Furthermore, as an application prospect, based on the suggested memristive neuron, we demonstrated the self-resting neural operation with the local circuit configuration and revealed probabilistic Bayesian inferences for genetic regulatory networks with low normalized mean squared errors (< ∼2.41 × 10-2 ) and its robustness to the ProbAct variation. This article is protected by copyright. All rights reserved.