[18p-P06-1~11] FS Focused Session "AI Electronics" (Poster)

[18p-P06-10] Comparison of magneto-optical diffractive deep neural networks and diffractive deep neural networks utilizing optical path differences

〇Juri Ikeda¹, Hotaka Sakaguchi¹, Hirofumi Nonaka², Hiroyuki Awano³, Fatima Zahra Chafi¹, Takayuki Ishibashi¹ (1.Nagaoka Univ. Tech, 2.Aichi Inst. Tech., 3.Toyota Tech. Inst.)

Keywords:

Magneto-Optical,neural network

Diffractive deep neural network (D2NN) is a type of stacked optical neural network that is expected to perform low-power and high-speed computations. We proposed a Magneto-Optical Diffractive Deep Neural Network (MO-D2NN) that utilizes the magneto-optical effect of magnetic materials. In this study, we compared and evaluated the performance of MO-D2NN with that of a previous study, D2NN using optical path difference.