(Invited) Neuromorphic Application of Oxide Semiconductors

Abstract

Artificial intelligences are promising as key technologies in future societies, but the conventional ones are complicated software executed on high-specked hardware. Therefore, we are investigating "brain-type integrated system", namely, neural network built only by hardware, which can be compact, low power, robust, and integrated on everything. In this study, we have succeeded in hardware simplification and are trying to utilize oxide semiconductors for the neuromorphic application because they can be fabricated using low cost fabrication such as sputtering and printing. We made a sandwiched structure of Ti/a-IGZO/Ti and found that the conductance decreases as the time goes by, which is available to modified Hebbian rule, a learning rule we proposed for our network. We modeled the conductance degradation of the a-IGZO thin film, developed a simulator, and obtained the learning results. We would like to develop brain-type integrated system with three-dimensional structure utilizing the oxide semiconductors in future.

Emerging applications using metal-oxide semiconductor thin-film devices

“…21 and 22. [98][99][100] In addition to the usage of metal-oxide semiconductor thin-film devices in neuromorphic systems itself, a local learning rule named modified Hebbian learning is proposed without additional circuits to control connection strength of synapse elements, which intentionally utilizes the conductance deterioration of the metal-oxide semiconductor thin-film devices. Such neuromorphic systems can be compact, because the control circuits are not necessary, which can also make the total system realized only by the metal-oxide semiconductor electronics, because metal-oxide semiconductor thin-film devices are not good at composing complicated circuits.…”

Switchover behavior between long-term potentiation and depression in amorphous Ga–Sn–O thin-film spike-timing-dependent-plasticity device

“…30) Therefore, it is believed that they are some of the most suitable materials for neuromorphic systems, which require distinctive properties and astronomically high integration. 31,32) In particular, we are investigating amorphous Ga-Sn-O (α-GTO) thin films, which do not contain In, a rare metal, 33,34) also for neuromorphic systems, 35,36) which require memristive devices [37][38][39][40][41][42][43][44][45] and STDP devices. 46,47) In this study, a switchover behavior between long-term potentiation (LTP) characteristic and long-term depression (LTD) characteristic in an α-GTO thin-film STDP device has been observed.…”