Figure 3 | Large-scale photonic in-memory computing. (IMAGE)

Light Publishing Center, Changchun Institute of Optics, Fine Mechanics And Physics, CAS

Figure 3 | Large-scale photonic in-memory computing.

Caption

Figure 3 | Large-scale photonic in-memory computing. a, 12 × 12 large-scale array of the trained LC vectorial DNN (upper) and its real-world photo under crossed polarizers (lower). b, Experimental results of large-scale photonic in-memory computing based on the LC device in (a), including: (i) optical input example “bad cab”; (ii) output intensity pattern after parallel optical computing; (iii) nonlinear relation between retrieved and recorded light; (iv) micrographs of another LC device after recording the computing results, with black or white arrows labeling the crossed polarizers and the yellow arrow labeling the optional full-wave plate; (v) light-speed optical retrieval of the computing results under crossed polarizers, which can be used in the next all-optical loops. The experimental results for the second input letter “a” are enlarged and labeled with dotted circles, with the ideal output polarization and LC orientations marked in (ii) and (iv), respectively.

Credit

Peng Chen et al.

Usage Restrictions

Credit must be given to the creator.

License

CC BY

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.