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

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.

Fig. 1

Integration of the h-BN/WSe₂ optic-neural synaptic device. a Schematic of the human optic nerve system, the h-BN/WSe₂ synaptic device integrated with h-BN/WSe₂ photodetector, and the simplified electrical circuit for the ONS device. Here, the light sources were dot lasers with wavelengths of 655 nm (red), 532 nm (green), and 405 nm (blue) with a fixed power density (P) of 6 mW cm⁻² for all wavelengths. b Excitatory and inhibitory postsynaptic current characteristics and extracted conductance changes of the h-BN/WSe₂ ONS device under different light conditions (no light and RGB). c Long-term potentiation and depression curves under different light conditions, where the synaptic device is controlled using input pulses with an amplitude of 0.3 V. d, e Nonlinearity magnitude (d) and the number of effective conductance states (threshold_ΔG = 0.3%) (e), which were extracted for different wavelengths

Fig. 2

Structure and operating mechanism of the h-BN/WSe₂ synaptic device. a Functional/structural/architectural comparison of biological synapse with our synthetic WSe₂/WCL/h-BN synaptic device. b X-TEM image of the WSe₂/WCL/h-BN structure, and the high-resolution images corresponding to the WSe₂/WCL and WCL/h-BN interfaces. c, d EELS (c) and EDS (d) mapping images obtained on the cross-section of the WSe₂/WCL/h-BN structure. e Current relaxation curves after pulse amplitudes of 0.1 V and 1 V, and contribution ratio from unrecovered electrons in fast traps and slow traps at 1 s after the pulse. f Illustration of energy band diagrams after pulse and after de-trapping of carriers in fast traps. g Change in postsynaptic conductance and the switching energy measured as a function of O₂ plasma process time. Here, all the V_pulse were applied with a duration of 10 ms

Fig. 3

Synaptic plasticity characterization of h-BN/WSe₂ synaptic device. a Long-term potentiation and depression characteristics by different input pulses with an amplitude of 0.3 V, 0.5 V, or 1 V. b Number of effective conductance states for the three cases with a threshold_ΔG = 0.3%, and LTP/LTD curves when 600 pulses are applied in each potentiation and depression. c Stability of conductance states with below 1% variation. d Spike-timing-dependent plasticity behavior obtained in the h-BN/WSe₂ synaptic device. The pre-spike and post-spike voltages are applied to the presynaptic and SCT, respectively

Fig. 4

Colored and color-mixed pattern recognition based on an artificial optic-neural network. a Developed ONN for recognition of 28 × 28 RGB-colored images. b Examples of the training and the testing datasets consisting of single-colored and color-mixed numeric pattern images, respectively. c Recognition rate as a function of number of training epochs. d Weight mapping images after the 12th and 600th training epoch. e Activation values of each output neuron in cases of a single-colored number (blue 4) and a color-mixed number (red/green-mixed 4) after the 600th training epoch