Presentation Information
[8p-N201-8]Evaluation of Nanoflower structure MoSe2 device for Physical Reservoir Computing
〇Koki Nietani1, Alif Syafiq Kamarol Zaman1, Yuki Usami1,2, Hirofumi Tanaka1,2 (1.LSSE, Kyushu Inst. Tech., 2.Neumorph Center, Kyushu Inst. Tech)
Keywords:
Reservoir Computing,Edge AI,2D materials
In this study, we evaluated a physical reservoir computing (RC) device based on molybdenum diselenide with a nanoflower structure (MoSe2 NF). The MoSe2 NF was deposited alone on a Si substrate, and I-V characteristics were measured across multiple electrodes. The results revealed strong nonlinearity and the presence of high-order harmonic components. In a waveform generation task, signals obtained from all output electrodes enabled accurate waveform reconstruction, and the evaluation of short-term memory capacity also demonstrated favorable retention performance. These findings suggest that MoSe2 NF is a highly promising material for high-performance RC devices.