Add to BibSonomyAbstract
We propose an all-photonic architecture to accelerate linear matrix processing by encoding information in the amplitudes of frequency states. Our design is unique in providing a unitary, reversible mode of computation at high speeds.
© 2022 The Author(s)
PDF Article | Presentation VideoMore Like This
Artificial Neural Network Training on an Optical Processor via Direct Feedback Alignment
Kilian Müller, Julien Launay, Iacopo Poli, Matthew Filipovich, Alessandro Capelli, Daniel Hesslow, Igor Carron, Laurent Daudet, Florent Krzakala, and Sylvain Gigan
jsiii_3_3 European Quantum Electronics Conference (EQEC) 2023
Design of dual-mode photonic crystal nanobeam cavity via artificial neural network
Zixing Gou, Zhe Han, Tongyu Nie, and Huiping Tian
JTu3B.33 CLEO: Applications and Technology (CLEO:A&T) 2022
High-speed time series prediction and classification on an all-optical neural network
Aashu Jha, Chaoran Huang, Hsuan-Tung Peng, Weipeng Zhang, Bhavin Shastri, and Paul R. Prucnal
Tu3G.3 Optical Fiber Communication Conference (OFC) 2022