[14p-K309-1~14] 3.4 Laser system and materials

2:15 PM - 2:30 PM JST(5:15 AM - 5:30 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[14p-K309-3] Autonomous optimisation of laser cavity using Deep Reinforcement Learning

〇Yuki Ikeya¹, Takashi Sekine¹, Shuntaro Tani², Toshio Otsu², Endo Tsubasa², Takaaki Morita¹, Yoshinori Tamaoki¹, Yoshinori Kato¹, Toshiyuki Kawashima¹, Yohei Kobayashi² (1.Hamamatsu Photonics K.K., 2.ISSP, Univ.Tokyo)

Keywords:

spatial light modulator,reinforcement learning,laser cavity

Although reinforcement learning is considered to be a suitable method for autonomous optimization of lasers in the development of laser oscillators, its applicability to real laser systems is unknown.
As one of the control methods for laser resonators, we considered using a spatial optical phase modulator (LCOS-SLM) instead of a mirror to dynamically and rapidly control the angle of the mirror and curvature of the lens, which have a high degree of freedom. The presentation will report the results of an attempt to maximize the output by autonomous control using deep reinforcement learning.