English

A Learning-Based Approach to Approximate Coded Computation

Information Theory 2022-05-23 v1 Machine Learning math.IT

Abstract

Lagrange coded computation (LCC) is essential to solving problems about matrix polynomials in a coded distributed fashion; nevertheless, it can only solve the problems that are representable as matrix polynomials. In this paper, we propose AICC, an AI-aided learning approach that is inspired by LCC but also uses deep neural networks (DNNs). It is appropriate for coded computation of more general functions. Numerical simulations demonstrate the suitability of the proposed approach for the coded computation of different matrix functions that are often utilized in digital signal processing.

Keywords

Cite

@article{arxiv.2205.09818,
  title  = {A Learning-Based Approach to Approximate Coded Computation},
  author = {Navneet Agrawal and Yuqin Qiu and Matthias Frey and Igor Bjelakovic and Setareh Maghsudi and Slawomir Stanczak and Jingge Zhu},
  journal= {arXiv preprint arXiv:2205.09818},
  year   = {2022}
}

Comments

Submitted to IEEE Information Theory Workshop (ITW) 2022