Self-Powered Organic Optoelectronic Synapses with Binarized Weights for Noise-Suppressed Visual Perception and High-Robustness Inference

Abstract

Neuromorphic optoelectrical synapses have shown great potential in edge artificial intelligence (AI) for energy-efficient sensory computing. However, environmental noise and device nonidealities pose immense challenges to the inference accuracy and robustness of neuromorphic devices and networks. Here, inspired by the hierarchical biological vision system, self-powered organic optoelectronic synaptic devices, which benefit from a simple asymmetric-electrode structure, are demonstrated. The multifunctional synapses can be configured as self-powered, denoising vision sensors for noise-suppressed visual perception and as photostimulated synapses for implementing a quantized neural network (QNN) by the differential synaptic structure. An ex situ training method by weight mapping from trained real-valued weights to binarized weights has been proposed to increase the robustness of organic synapses. The binary neural network (BNN) with binarized weights mitigates the impact of the synaptic nonlinearity with 97.0% inference accuracy and exhibits only 1.7% accuracy reduction with large synaptic weight variations due to device nonidealities compared with the accuracy using backpropagation algorithms. Therefore, this self-powered organic synapse enables simultaneous optical sensing and processing for more robust and energy-efficient neuromorphic systems.

Modulating Neuromorphic Behavior of Organic Synaptic Electrolyte-Gated Transistors Through Microstructure Engineering and Potential Applications

“…During stimulation of the first spike, holes accumulate near the P3HT/ion–gel dielectric interface to form a conducting channel, resulting in the occurrence of EPSC of the three kinds of devices. In the SS device, some holes could be captured by PMMA molecules nearby the conducting channel during the first spike. ,,, During the stimulation of the second spike, in the ST and the DS devices, more holes further accumulate in the conducting channel, causing a higher PSC than that under the first spike (PPF behavior, Figure b). Furthermore, a larger PSC increment under the second spike, namely, a higher PPR (Figure c), can be observed in the DS device compared with the ST device because a rougher P3HT/ion–gel dielectric interface can generate a strong EDL to enhance hole accumulation and better crystalline quality can lead to efficient hole transport.…”

“…Furthermore, organic/polymer semiconductors exhibit superior biocompatibility compared to inorganic semiconductors, making them attractive for bioelectronic device applications. Consequently, numerous studies have explored the use of organic/polymer semiconductors in synaptic transistors that mimic biological synapses, − enabling new possibilities in neuromorphic computing, artificial intelligence, and brain–machine interfaces.…”

In-sensor neuromorphic computing using perovskites and transition metal dichalcogenides

“…Although single-device based synaptic elements show great potential in perceptual recognition, their recognition accuracy will be greatly limited in the face of interference from external mechanical simulations. Simultaneously, the information processing capacity of a single device is limited [273], and array-based processing can process more information even in the face of external interference [274], so it is particularly important to study the array-level in-sensor computing. For example, Seo et al prepared a flexible van der Waals photo synaptic array unit consisting of 25 optoelectronic synaptic devices based on ReS 2 and hexagonal boron nitride heterostructures [231].…”

Logical Perception of Artificial Vision Based on Nonlinear Neuromorphic Responses of Optoelectronic Synapses

“…Conventional systems of artificial visual perception are based on von Neumann architecture, which suffers from colossal energy consumption due to the separation of data storage and computing units . Inspired by the operating mechanism of the human brain, neuromorphic computing based on optoelectronic synapses has been intensively explored. − Under optical stimulation, the optoelectronic synapses trigger persistent electronic responses including excitatory postsynaptic currents and inhibitory postsynaptic currents. In previous studies, excitatory postsynaptic current and inhibitory postsynaptic current can be realized by electrical or optical pathways, demonstrating memory behaviors including short-term/long-term memory, − voltage-modulated plasticity, , paired-pulse facilitation (PPF), and learning ability .…”