Manifold-Aware General Coded Computing for Straggler-Resilient Distributed Computing
Abstract
Existing coded-computing designs do not explicitly exploit the intrinsic structure of the input data. In communication systems, statistical structure and redundancy are often removed through source coding (or compression) before channel coding is applied. This principle, however, does not transfer directly to coded computation. In many computational tasks, particularly in machine learning, the structure of the data is precisely what the computation seeks to exploit to infer outputs or learn meaningful patterns. Consequently, coded-computing schemes should preserve and leverage this structure in their code design, rather than ignoring or eliminating it through source coding. This observation motivates a different perspective on code construction. In many channel-coding schemes, such as Reed-Solomon codes, coded symbols are generated by evaluating a low-dimensional algebraic representation at selected points. In contrast, many high-dimensional datasets naturally concentrate near low-dimensional manifolds. In this paper, we exploit this intrinsic geometry by designing coded samples that follow the natural manifold of the data, rather than imposing an artificial low-dimensional structure unrelated to the data distribution. Inspired by graph-based manifold learning, we propose a manifold-aware encoding strategy for general coded computing (GCC). Experiments on neural network inference and high-dimensional polynomial evaluation demonstrate that the proposed strategy consistently and significantly reduces the mean squared recovery error under straggling compared with standard GCC.
Related
- Entanglement-Capability-Aware Finite-Bond Circuit Cutting for Distributed Quantum Computing
- Computing key rates for one-sided device-independent quantum key distribution
- Computing Smooth Geodesics under Two-Sided Curvature Bounds with Applications to Robotics and Image Analysis
- The complexity of computing the covering radius of a Euclidean lattice
- On computing deformations of contraction algebras of type E
BibTeX
@misc{moradi2026manifoldaware,
title = {{Manifold-Aware General Coded Computing for Straggler-Resilient Distributed Computing}},
author = {Parsa Moradi and Mohammad Ali Maddah-Ali},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.00552},
eprint = {2609.00552},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.00552},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Manifold-Aware General Coded Computing for Straggler-Resilient Distributed Computing
AU - Parsa Moradi
AU - Mohammad Ali Maddah-Ali
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.00552
UR - https://arxiv.org/abs/2609.00552
AB - Existing coded-computing designs do not explicitly exploit the intrinsic structure of the input data. In communication systems, statistical structure and redundancy are often removed through source coding (or compression) before channel coding is applied. This principle, however, does not transfer directly to coded computation. In many computational tasks, particularly in machine learning, the structure of the data is precisely what the computation seeks to exploit to infer outputs or learn meaningful patterns. Consequently, coded-computing schemes should preserve and leverage this structure in their code design, rather than ignoring or eliminating it through source coding. This observation motivates a different perspective on code construction. In many channel-coding schemes, such as Reed-Solomon codes, coded symbols are generated by evaluating a low-dimensional algebraic representation at selected points. In contrast, many high-dimensional datasets naturally concentrate near low-dimensional manifolds. In this paper, we exploit this intrinsic geometry by designing coded samples that follow the natural manifold of the data, rather than imposing an artificial low-dimensional structure unrelated to the data distribution. Inspired by graph-based manifold learning, we propose a manifold-aware encoding strategy for general coded computing (GCC). Experiments on neural network inference and high-dimensional polynomial evaluation demonstrate that the proposed strategy consistently and significantly reduces the mean squared recovery error under straggling compared with standard GCC.
ER -