Artificial optic-neural synapse for colored and color-mixed pattern recognition

TLDR

Synaptic device research has focused on demonstrating synaptic dynamics emulation rather than functionalizing devices for complex learning. The authors aim to demonstrate an optic‑neural synaptic device that integrates synaptic and optical‑sensing functions on an h‑BN/WSe₂ heterostructure, advancing neural networks that combine sensing and training for complex pattern recognition. The device is fabricated on an h‑BN/WSe₂ heterostructure that simultaneously provides synaptic behavior and optical sensing. When arranged in an optic‑neural network, the device accurately and energy‑efficiently recognizes colored and color‑mixed patterns, exhibiting a near‑linear weight‑update trajectory, many stable conduction states with <1 % variation, and operating with 0.3 V spikes at 66 fJ per spike.

Abstract

The priority of synaptic device researches has been given to prove the device potential for the emulation of synaptic dynamics and not to functionalize further synaptic devices for more complex learning. Here, we demonstrate an optic-neural synaptic device by implementing synaptic and optical-sensing functions together on h-BN/WSe2 heterostructure. This device mimics the colored and color-mixed pattern recognition capabilities of the human vision system when arranged in an optic-neural network. Our synaptic device demonstrates a close to linear weight update trajectory while providing a large number of stable conduction states with less than 1% variation per state. The device operates with low voltage spikes of 0.3 V and consumes only 66 fJ per spike. This consequently facilitates the demonstration of accurate and energy efficient colored and color-mixed pattern recognition. The work will be an important step toward neural networks that comprise neural sensing and training functions for more complex pattern recognition.