possible due to the parallel computing that efficiently processes and memorizes the information at the same time through neural network that consists of ≈10 12 neurons and ≈10 15 synapses. [6][7][8][9] This allows the brain to be efficient in handling cognitive operations such as think, read, learn, remember, and reason. [10][11][12][13] Therefore, in order to mimic the brain, neuromorphic computing that simulates the neural network has been developed and in considerable attention ever since. As one of the neuromorphic applications, neuromorphic visual system, i.e., optical synapse, that mimics the biological visual system is one of the most researched due to its importance as the development of artificial eyes would benefit humans greatly. However, this neuromorphic visual system consists of a separate photosensor and a neuromorphic synapse that are connected via a circuitry, leading to a low efficiency and a complex circuitry. Although the neuromorphic device enables processing and memorizing in one device, the signal itself should be converted to electric signal by the photodetector and transported through the circuit to the neuromorphic device. This extra transportation of the signal requires extra energy consumption and also could cause the bottleneck effect similar to von-Neumann computing system, diminishing the advantages of using neuromorphic devices. [14][15][16][17] Therefore, in order to improve the visual information processing efficiency, optical synapses are also developed where it has both the synaptic functions and the photo-sensing abilities. [18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33] Different types of devices, such as memristors, field-effect transistors, and phase change memories, have been studied for the application of artificial synapses. [16,18,19,[24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41]44] Additionally, numerous materials including but not limited to low-dimensional materials, perovskites, oxide semiconductors, and organic materials have been applied to artificial synapses. [16,18,19,[24][25][26][27][28][29][30][31][32][41][42][43][44] Among these, oxide semiconductor thin-film transistors show promising properties for optical synapses. [18,19,24,[27][28][29][30][31][32] Oxide semiconductors, in terms of light detection, have wavelength-and intensity-selectivity as the current increases when intensity increases and wavelength decreases. This property is crucial to optical neuromorphic synapses for learning property.