A Reconfigurable Optoelectronic Synaptic Transistor with Stable Zr‐CsPbI₃ Nanocrystals for Visuomorphic Computing (Adv. Mater. 12/2023)

Abstract

Synaptic Transistors In article 2208497, Haifeng Ling, Linghai Xie, Wei Huang, and co‐workers propose a wavelength‐aware reconfigurable synaptic transistor for visuomorphic computing. Based on the broadband absorption of Zr‐CsPbI3 perovskite nanocrystals, reconfigurability between short‐term plasticity and long‐term plasticity for the device is realized. A vision sensor array is constructed to simulate image pre‐processing and recognition with a homogeneous architecture.

Weak Light‐Stimulated Synaptic Transistors Based on MoS2/Organic Semiconductor Heterojunction for Neuromorphic Computing

“…Similarly, the conversion could also be achieved by applying optical pulses with different frequencies and numbers (Figure 2h,i). To further evaluate the conversion process from STM to LTM, a typical stretched exponential function was introduced to fit the decay curves: [54,55]…”

“…Similarly, the conversion could also be achieved by applying optical pulses with different frequencies and numbers (Figure 2h,i). To further evaluate the conversion process from STM to LTM, a typical stretched exponential function was introduced to fit the decay curves: [ 54,55 ] $$$$\begin{eqnarray}I\; =\; ({I}_{\rm{0}}{\rm{-}}{I}_\infty {\rm{)\ exp}}\left[ {{\rm{ - }}{{\left( {\frac{{t{\rm{ - }}{t}_{\rm{0}}}}{{{\tau}}}} \right)}}^{{\beta}}} \right]{\rm{ + }}\,{I}_\infty \end{eqnarray}$$$$where I 0 , t 0 , τ , β , and I ∞ denoted the maximum photocurrent, the ending time of the optical pulse, the decay time, the stretch index between 0 and 1, and the background current, respectively. Figure S6 (Supporting Information) displays the gradual increase of τ with the increase of width, frequency, and number of the optical pulses, further demonstrating the conversion process from STM to LTM.…”

Optoelectronic Synapses and Photodetectors Based on Organic Semiconductor/Halide Perovskite Heterojunctions: Materials, Devices, and Applications

“…[ 119,120 ] Significant efforts have been made to construct artificial visual systems based on OSDs for mimicking visual perception functions, which can be a crucial advance towards robotic vision and artificial intelligence systems. [ 121–123 ] The development of OSDs based on organic semiconductor/halide perovskite heterojunctions for artificial visual systems promises to facilitate the development of ANNs and biomimetic eye technologies that can implement intelligent pattern recognition, preprocessing, visual receptor, and image memory functions. The performance comparison of various types of OSDs is presented in Table 2 .…”

Hierarchies in Visual Pathway: Functions and Inspired Artificial Vision

“…[80] Contrary to conventional CMOS image sensors, neuromorphic devices have been demonstrated to show tunable volatility and complex dynamics, ensuring accurate simulation of biological retinal behaviors. [81,82] Actually, the responsivity of independent elements modulated by external stimuli is designed to emulate the change of synaptic weight during the learning process. The interfacing and data transferring between the traditional vision sensor and processing units can be eliminated by the photonic synaptic device array where the external light stimuli can be expressed as an S vector to multiply responsivity vector R within a single clock cycle.…”