Design automation of photonic resonator weights.

Autor:	Ferreira de Lima T; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Doris EA; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Bilodeau S; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Zhang W; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Jha A; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Peng HT; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Blow EC; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA., Huang C; Department of Electrical Engineering, The Chinese University of Hong Kong, Hong Kong SAR, China., Tait AN; Department of Electrical and Computer Engineering, Queen's University, Kingston K7L 3N6, ON, Canada., Shastri BJ; Department of Physics, Engineering Physics & Astronomy, Queen's University, Kingston K7L 3N6, ON, Canada.; Vector Institute of Artificial Intelligence, Toronto MS1 5G1, ON, Canada., Prucnal PR; Department of Electrical and Computer Engineering, Princeton University, Princeton 08544, NJ, USA.
Jazyk:	angličtina
Zdroj:	Nanophotonics (Berlin, Germany) [Nanophotonics] 2022 Apr 11; Vol. 11 (17), pp. 3805-3822. Date of Electronic Publication: 2022 Apr 11 (Print Publication: 2022).
DOI:	10.1515/nanoph-2022-0049
Abstrakt:	Neuromorphic photonic processors based on resonator weight banks are an emerging candidate technology for enabling modern artificial intelligence (AI) in high speed analog systems. These purpose-built analog devices implement vector multiplications with the physics of resonator devices, offering efficiency, latency, and throughput advantages over equivalent electronic circuits. Along with these advantages, however, often come the difficult challenges of compensation for fabrication variations and environmental disturbances. In this paper, we review sources of variation and disturbances from our experiments, as well as mathematically define quantities that model them. Then, we introduce how the physics of resonators can be exploited to weight and sum multiwavelength signals. Finally, we outline automated design and control methodologies necessary to create practical, manufacturable, and high accuracy/precision resonator weight banks that can withstand operating conditions in the field. This represents a road map for unlocking the potential of resonator weight banks in practical deployment scenarios. (© 2022 the author(s), published by De Gruyter, Berlin/Boston.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu