Artificial intelligence is a promising concept in modern and future societies. Presently, software programs are used but with a bulky computer size and large power consumption. Conversely, hardware systems named neuromorphic systems are suggested, with a compact computer size and low power consumption. An important factor is the number of processing elements that can be integrated. In the present study, three decisive technologies are proposed: (1) amorphous metal oxide semiconductor thin films, one of which, Ga–Sn–O (GTO) thin film, is used. GTO thin film does not contain rare metals and can be deposited by a simple process at room temperature. Here, oxygen-poor and oxygen-rich layers are stacked. GTO memristors are formed at cross points in a crossbar array; (2) analog memristor, in which, continuous and infinite information can be memorized in a single device. Here, the electrical conductance gradually changes when a voltage is applied to the GTO memristor. This is the effect of the drift and diffusion of the oxygen vacancies (Vo); and (3) autonomous local learning, i.e., extra control circuits are not required since a single device autonomously modifies its own electrical characteristic. Finally, a neuromorphic system is assembled using the abovementioned three technologies. The function of the letter recognition is confirmed, which can be regarded as an associative memory, a typical artificial intelligence application.
A memristor property of an amorphous Sn–Ga–O (α-TGO) thin-film device deposited using a mist chemical-vapor-deposition (mist-CVD) method has been found. The α-TGO device can be manufactured at a low cost because it does not include rare metals such as In. Moreover, it is expected that the α-TGO device can be manufactured at an even lower cost because the mist-CVD method is performed at atmospheric pressure. Here, the α-TGO layer was deposited using a hot-wall-type mist-CVD method. The hysteresis curve of the memristor characteristic was certainly obtained, and the electric resistances for the high- and low-resistance states were stably repeated at least 20 times. Although the switching ratio and repeatability are not sufficient in the case that it is applied to resistive random access memories, they are acceptable for some applications such as synapse elements in neuromorphic systems.
As the amount of data that people handle increases, the conventional Neumann-type computer architecture is reaching its limits. Therefore, research on hardware implementation of machine learning systems is being actively conducted. In this paper, we have implemented and evaluated neuromorphic hardware that realizes human brain neurons and synapses using oxide semiconductor of amorphous In-Ga-Zn-O (a-IGZO) and a cellular neural network. It was confirmed how variations of initial resistance and deterioration rate of the oxide semiconductor affect operation accuracy of the neuromorphic hardware. Furthermore, we clarified that an activation function suitable for the hardware implementation is a ReLU function.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.