Hyperspectral In-Memory Computing Using Frequency Comb and Programmable Optical Memories
We demonstrate a hyperspectral compute-in-memory architecture that performs single-shot matrix-matrix multiplication using optical frequency combs and programmable memories, leveraging space and wavelength multiplexing for dense, energy-efficient optical acceleration of AI workloads.
Authors
- Mostafa Honari Latifpour
- Myoung‐Gyun Suh (ORCID: https://orcid.org/0000-0002-9527-0585)
- Byoung Jun Park
- Y. Yamamoto (ORCID: https://orcid.org/0000-0002-4150-6804)
Publication Details
- Published
- 2025-06-29
- DOI
- https://doi.org/10.23919/oecc/psc62146.2025.11109537
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00