Grid-like Error-Correcting Codes for Matrix Multiplication with Better Correcting Capability

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Hao, Jiang, Zhengyi, Huang, Zhongyi, Bai, Bo, Zhang, Gong, Hou, Hanxu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911094515171328
author Shi, Hao
Jiang, Zhengyi
Huang, Zhongyi
Bai, Bo
Zhang, Gong
Hou, Hanxu
author_facet Shi, Hao
Jiang, Zhengyi
Huang, Zhongyi
Bai, Bo
Zhang, Gong
Hou, Hanxu
contents Matrix multiplication over the real field constitutes a foundational operation in the training of deep learning models, serving as a computational cornerstone for both forward and backward propagation processes. However, the presence of silent data corruption (SDC) in large-scale distributed training environments poses a significant threat to model convergence and predictive accuracy, particularly when such errors manifest during matrix multiplication. Due to their transient and non-intrusive nature, these errors often evade detection, allowing them to propagate and accumulate over time, ultimately leading to substantial degradation in model performance. In this paper, we introduce a novel error-correcting coding framework specifically tailored for matrix multiplication operations. Our proposed framework is designed to detect and correct multiple computational errors that may arise during the execution of matrix products. By leveraging a grid-based structural encoding scheme, our approach enhances error localization and correction capabilities across all participating matrices, thereby significantly improving the fault tolerance of the computation. Experimental results demonstrate that our method achieves deterministic correction of up to two erroneous symbols distributed across three matrices with 100\% reliability, while incurring only a 24\% overhead in computational time on GPU architectures. Furthermore, we provide a rigorous theoretical analysis of the error-correction properties inherent to our coding scheme, establishing its correctness and robustness under well-defined fault models.
format Preprint
id arxiv_https___arxiv_org_abs_2508_04355
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Grid-like Error-Correcting Codes for Matrix Multiplication with Better Correcting Capability
Shi, Hao
Jiang, Zhengyi
Huang, Zhongyi
Bai, Bo
Zhang, Gong
Hou, Hanxu
Information Theory
Matrix multiplication over the real field constitutes a foundational operation in the training of deep learning models, serving as a computational cornerstone for both forward and backward propagation processes. However, the presence of silent data corruption (SDC) in large-scale distributed training environments poses a significant threat to model convergence and predictive accuracy, particularly when such errors manifest during matrix multiplication. Due to their transient and non-intrusive nature, these errors often evade detection, allowing them to propagate and accumulate over time, ultimately leading to substantial degradation in model performance. In this paper, we introduce a novel error-correcting coding framework specifically tailored for matrix multiplication operations. Our proposed framework is designed to detect and correct multiple computational errors that may arise during the execution of matrix products. By leveraging a grid-based structural encoding scheme, our approach enhances error localization and correction capabilities across all participating matrices, thereby significantly improving the fault tolerance of the computation. Experimental results demonstrate that our method achieves deterministic correction of up to two erroneous symbols distributed across three matrices with 100\% reliability, while incurring only a 24\% overhead in computational time on GPU architectures. Furthermore, we provide a rigorous theoretical analysis of the error-correction properties inherent to our coding scheme, establishing its correctness and robustness under well-defined fault models.
title Grid-like Error-Correcting Codes for Matrix Multiplication with Better Correcting Capability
topic Information Theory
url https://arxiv.org/abs/2508.04355