Reconfigurable nonlinear photonic activation function for photonic neural network based on non-volatile opto-resistive RAM switch

Abstract

Photonic neural network has been sought as an alternative solution to surpass the efficiency and speed bottlenecks of electronic neural network. Despite that the integrated Mach–Zehnder Interferometer (MZI) mesh can perform vector-matrix multiplication in photonic neural network, a programmable in-situ nonlinear activation function has not been proposed to date, suppressing further advancement of photonic neural network. Here, we demonstrate an efficient in-situ nonlinear accelerator comprising a unique solution-processed two-dimensional (2D) MoS2 Opto-Resistive RAM Switch (ORS), which exhibits tunable nonlinear resistance switching that allow us to introduce nonlinearity to the photonic neuron which overcomes the linear voltage-power relationship of typical photonic components. Our reconfigurable scheme enables implementation of a wide variety of nonlinear responses. Furthermore, we confirm its feasibility and capability for MNIST handwritten digit recognition, achieving a high accuracy of 91.6%. Our accelerator constitutes a major step towards the realization of in-situ photonic neural network and pave the way for the integration of photonic integrated circuits (PIC).

1T MoS2/CoS2 heterostructures enabling enhanced resistive switching behavior in sodium alginate-based flexible memristors

“…63,64 As shown in Fig. 7c, when the positive bias voltage applied to the top electrode is increased, the fitted slope is approximated to 2 (I p V 2 ), indicating that more electrons are injected into the interlayer to be trapped by the V S and migrate to form CFs, 65 and the carrier transport is dominated by the SCLC mechanism. As time goes by, the V S is filled with the charge, and electrons can jump between V S , 66 and the steep current increase (I p V n ) occurs to switch the device from HRS to LRS to complete the ''set'' process.…”

Neuromorphic Photonics Circuits: Contemporary Review

“…Since achieving nonlinearity of the characteristic only on photonic elements is a challenging task, many researchers are developing various combinations of photonic elements that can influence the characteristic for the necessary adjustment by electronic components. Also, solutions include using hybrid structures (Ge/Si hybrid structure in a micro-ring resonator) [76], structures using the free-carrier dispersion effect (scheme with a Mach-Zehnder interferometer loaded with MCR, heating elements, and a Mach-Zehnder coupler) [76][77][78] and another popular direction-phase change material (PCM) coatings [79,80]. The given examples can realize not one but several variants of activation functions: radial basis, sigmoid, softplus, Relu, and ELU.…”

Solution-Processable and Printable Two-Dimensional Transition Metal Dichalcogenide Inks

“…Similarly, MoS 2 was also tested for photonic neural networks. Xu et al proposed an optical-to-optical nonlinear activation accelerator by bridging a unique solution-processed 2D MoS 2 light-sensitive opto-resistive RAM switch to the photonic artificial neural network . The accelerator is an important step toward realizing in situ photonic neural networks and the integration of photonic integrated circuits (PIC).…”