Summary
2023
Session Number:C4L-3
Session:
Number:C4L-34
Approximate Reservoir Computing Using Semiconductor Lasers with Optical Feedback
Ito Tatsuki, Kanno Kazutaka, Uchida Atsushi,
pp.629-632
Publication Date:2023-09-21
Online ISSN:2188-5079
Summary:
We propose a concept of approximate reservoir computing by combine approximate computing with reservoir com-puting. We use a semiconductor laser with optical feed-back as a nonlinear device for reservoir computing. We quantize the virtual node states and the output weights for reservoir computing, and perform a time-series prediction task. We found that the performance of time-series predic-tion is degraded when the number of bits for quantization are reduced.