Integrated non-reciprocal magneto-optics with ultra-high endurance for photonic in-memory computing
Abstract Processing information in the optical domain promises advantages in both speed and energy efficiency over existing digital hardware for a variety of emerging applications in artificial intelligence and machine learning. A typical approach to photonic processing is to multiply a rapidly changing optical input vector with a matrix of fixed optical weights. However, encoding these weights on-chip using an array of photonic memory cells is currently limited by a wide range of material- and device-level issues, such as the programming speed, extinction ratio and endurance, among others. Here we propose a new approach to encoding optical weights for in-memory photonic computing using magneto-optic memory cells comprising heterogeneously integrated cerium-substituted yttrium iron garnet (Ce:YIG) on silicon micro-ring resonators. We show that leveraging the non-reciprocal phase shift in such magneto-optic materials offers several key advantages over existing architectures, providing a fast (1 ns), efficient (143 fJ per bit) and robust (2.4 billion programming cycles) platform for on-chip optical processing.
Authors
- Vivswan Shah (ORCID: https://orcid.org/0000-0003-1786-705X)
- Toshiya Murai (ORCID: https://orcid.org/0000-0002-0990-7384)
- Duanni Huang (ORCID: https://orcid.org/0000-0002-5873-4141)
- Yuya Shoji (ORCID: https://orcid.org/0000-0003-1907-5136)
- Paolo Pintus (ORCID: https://orcid.org/0000-0002-6968-7483)
- Nathan Youngblood (ORCID: https://orcid.org/0000-0003-2552-9376)
- Mario Dumont (ORCID: https://orcid.org/0000-0002-3956-612X)
- Galan Moody (ORCID: https://orcid.org/0000-0002-6265-2034)
- John E. Bowers (ORCID: https://orcid.org/0000-0003-4270-8296)
Institutions
- Tokyo Institute of Technology (JP)
- University of California, Santa Barbara (US)
- University of Pittsburgh (US)
- University of Cagliari (IT)
- National Institute of Advanced Industrial Science and Technology (JP)
Publication Details
- Journal
- Nature Photonics
- Published
- 2024-10-23
- DOI
- https://doi.org/10.1038/s41566-024-01549-1
- Citations
- 53
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 18.41
Funders
- National Science Foundation
- University of Pittsburgh
- Università degli Studi di Cagliari
- Fondazione di Sardegna
- Regione Autonoma della Sardegna
- Directorate for Computer and Information Science and Engineering
- U.S. Air Force
- Center for Research Computing, University of Pittsburgh
- Division of Electrical, Communications and Cyber Systems