Creation of brain-inspired material intelligence composed of nanomaterial network

2021 Fiscal Year Final Research Report

• Project/Area Number: 20K22485
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0402:Nano/micro science, applied condensed matter physics, applied physics and engineering, and related fields
• Research Institution: Kyushu Institute of Technology
• Principal Investigator: Yuki Usami 九州工業大学, 大学院生命体工学研究科, 助教 (60878437)
• Project Period (FY): 2020-09-11 – 2022-03-31
• Keywords: リザバー演算 / 光応答素子 / 金属酸化物

Outline of Final Research Achievements

In this study, we implemented reservoir computing using α-Fe2O3/Ti-Bi-O nanoparticle composites, which are highly photo-responsive materials in order to perform reservoir computing driven by fluctuations. Structural analysis results showed the formation of nanoparticles and revealed that the current value increased upon light irradiation. From impedance spectroscopy, ionic conduction at the interface was dominant in this material system. Furthermore, the results of the waveform generation task, a representative benchmark, showed that a certain amount of light irradiation increased the accuracy of waveform prediction by optimizing the amount of current noise. These results show that fluctuations caused by light irradiation successfully control the reservoir computing performance.

Free Research Field

材料科学

Academic Significance and Societal Importance of the Research Achievements

本研究を通して、光応答材料を演算部に組み込むことで、外部からの光刺激により電流ノイズの量が最適化され、演算機能が制御可能となることを見出した。本研究の進展により、ノイズや揺らぎといったこれまで演算機能向上の妨げとなっていたものを逆に活用する、真に脳機能を模倣した演算処理の実現が期待される。