Reproducible Ozaki‐Based Matrix Multiplication on Photonic Matrix Processors With Algorithm‐Based Fault Tolerance
Open Access
- 10 August 2026
- journal article
- research article
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 38 (16) , e70895
Abstract
Photonic matrix processors are attracting attention in applications such as AI inference due to their low‐power and low‐latency properties. However, their applicability has been largely limited to low‐precision computation due to inherent stochastic analog noise and device non‐idealities. This limitation creates a large gap between the capabilities of photonic matrix processors and the potential demand for high‐precision computation in scientific computing. In this study, we propose a method to reproduce the results of the Ozaki‐based high‐precision matrix multiplication on low‐precision photonic matrix processors by combining the Ozaki scheme with a simple Algorithm‐Based Fault Tolerance (ABFT). The Ozaki scheme decomposes high‐precision operations into multiple low‐precision computations, while the ABFT introduces algorithm‐level redundancy to detect and correct errors caused by stochastic analog noise. Evaluation results show that ABFT significantly reduces failure probability by orders of magnitude in low‐error regimes, despite the increased number of sub‐computations introduced by the Ozaki decomposition. Furthermore, the overhead of ABFT remains small compared to the total number of photonic tensor‐core invocations, enabling high‐precision computation across a wide range of matrix sizes.Keywords
Funding Information
- Japan Society for the Promotion of Science (JP22H05193)
- Japan Science and Technology Agency (JPMJCR24R3)
This publication has 12 references indexed in Scilit:
- Achieving high precision in analog in-memory computing systemsnpj Unconventional Computing, 2026
- Universal photonic artificial intelligence accelerationNature, 2025
- An integrated large-scale photonic accelerator with ultralow latencyNature, 2025
- Photonic matrix multiplication lights up photonic accelerator and beyondLight: Science & Applications, 2022
- Silicon Photonics for Artificial Intelligence Acceleration : HotChips 322020
- DGEMM Using Tensor Cores, and Its Accurate and Reproducible Versions2020
- Algorithm-Based Fault Tolerance for Dense Matrix Factorizations, Multiple Failures and AccuracyACM Transactions on Parallel Computing, 2015
- Fault tolerant matrix-matrix multiplication2011
- Error-free transformations of matrix multiplication by using fast routines of matrix multiplication and its applicationsNumerical Algorithms, 2011
- Algorithm-Based Fault Tolerance for Matrix OperationsIEEE Transactions on Computers, 1984