Programmable chalcogenide-based all-optical deep neural networks

Abstract

We demonstrate a passive all-chalcogenide all-optical perceptron scheme. The network's nonlinear activation function (NLAF) relies on the nonlinear response of Ge₂Sb₂Te₅ to femtosecond laser pulses. We measured the sub-picosecond time-resolved optical constants of Ge₂Sb₂Te₅ at a wavelength of 1500 nm and used them to design a high-speed Ge₂Sb₂Te₅-tuned microring resonator all-optical NLAF. The NLAF had a sigmoidal response when subjected to different laser fluence excitation and had a dynamic range of -9.7 dB. The perceptron's waveguide material was AlN because it allowed efficient heat dissipation during laser switching. A two-temperature analysis revealed that the operating speed of the NLAF is <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mrow><mml:mo>≤</mml:mo> <mml:mn>1</mml:mn></mml:mrow> </mml:math> ns. The percepton's nonvolatile weights were set using low-loss Sb₂S₃-tuned Mach Zehnder interferometers (MZIs). A three-layer deep neural network model was used to test the feasibility of the network scheme and a maximum training accuracy of 94.5% was obtained. We conclude that combining Sb₂S₃-programmed MZI weights with the nonlinear response of Ge₂Sb₂Te₅ to femtosecond pulses is sufficient to perform energy-efficient all-optical neural classifications at rates greater than 1 GHz.