Controllable SiOx Nanorod Memristive Neuron for Probabilistic Bayesian Inference. Issue 1 (22nd October 2021)

• Record Type: Journal Article
• Authors: Choi, Sanghyeon; Kim, Gwang Su; Yang, Jehyeon; Cho, Haein; Kang, Chong‐Yun; Wang, Gunuk
• Is Part Of: Advanced materials. Volume 34:Issue 1(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-10-22
• Subjects: artificial neurons -- memristors -- nanorods -- neuromorphic computing -- probabilistic neural networks -- silicon oxide; Materials -- Periodicals; Chemical vapor deposition -- Periodicals; 620.11
• DOI: 10.1002/adma.202104598
• Languages: English
• ISSNs: 0935-9648

Abstract

Abstract: Modern artificial neural network technology using a deterministic computing framework is faced with a critical challenge in dealing with massive data that are largely unstructured and ambiguous. This challenge demands the advances of an elementary physical device for tackling these uncertainties. Here, we designed and fabricated a SiO x 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 nanorod structure provides the random distribution of multiple nanopores all across the active area, capable of forming a multitude of Si filaments at many SiO x nanorod edges after the electromigration process, leading to a stochastic switching event with very high dynamic range (≈5.15 × 10 10 ) and low energy (≈4.06 pJ). 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. Abstract : A probabilistic SiO x nanorod memristive neuron is fabricated by employing the glancing angle deposition technique. Due to the SiO x nanorod structure, a multitude of Si switching filaments at many SiO x nanorod edges can be formed, leading to different probabilistic activation functions with controllability. Furthermore, the self‐resting neural operation and probabilistic Bayesian inferences are also demonstrated with the fabricated device.