Explore Neuromorphic Systems by Integrating AOS thin-film Synapse devices with LSI circuits for enhanced AI performance

Neuromorphic chip integrated with a large-scale integration circuit and amorphous-metal-oxide semiconductor thin-film synapse devices

Artificial intelligence is a promising technology for future societies. Neural networks have many advantages, including self-organization and self-learning. They also offer parallel distributed computing and fault tolerance. However, their size and energy consumption are high. The hardware of neuromorphic systems is similar to that of living brains. They have the same benefits, such as compact size, low-power, and robust operation. However, some systems are not optimized, which means they only gain a portion. For example, machine-learning may be processed elsewhere in order to download fixed parameters. We are investigating neuromorphic systems in different ways to solve these problems. In this study, a neuromorphic chip integrated with a large-scale integration circuit (LSI) and amorphous-metal-oxide semiconductor (AOS) thin-film synapse devices has been developed.

Source:
https://www.nature.com/articles/s41598-022-09443-y