Distributed Approximate Computing with Constant Locality

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yuan, Deheng, Guo, Tao, Huang, Zhongyi, Jin, Shi
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913247466094592
author Yuan, Deheng
Guo, Tao
Huang, Zhongyi
Jin, Shi
author_facet Yuan, Deheng
Guo, Tao
Huang, Zhongyi
Jin, Shi
contents Consider a distributed coding for computing problem with constant decoding locality, i.e., with a vanishing error probability, any single sample of the function can be approximately recovered by probing only constant number of compressed bits. We establish an achievable rate region by designing an efficient layered coding scheme, where the coding rate is reduced by introducing auxiliary random variables and local decoding is achieved by exploiting the expander graph code. Then we show the rate region is optimal under mild regularity conditions on source distributions. The proof relies on the reverse hypercontractivity and a rounding technique to construct auxiliary random variables. The rate region is strictly smaller than that for the classical problem without the constant locality constraint in most cases, which indicates that more rate is required in order to achieve lower coding complexity. Moreover, a coding for computing problem with side information is analogously studied. We also develop graph characterizations, which simplifies the computation of the achievable rate region.
format Preprint
id arxiv_https___arxiv_org_abs_2312_04141
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Distributed Approximate Computing with Constant Locality
Yuan, Deheng
Guo, Tao
Huang, Zhongyi
Jin, Shi
Information Theory
Consider a distributed coding for computing problem with constant decoding locality, i.e., with a vanishing error probability, any single sample of the function can be approximately recovered by probing only constant number of compressed bits. We establish an achievable rate region by designing an efficient layered coding scheme, where the coding rate is reduced by introducing auxiliary random variables and local decoding is achieved by exploiting the expander graph code. Then we show the rate region is optimal under mild regularity conditions on source distributions. The proof relies on the reverse hypercontractivity and a rounding technique to construct auxiliary random variables. The rate region is strictly smaller than that for the classical problem without the constant locality constraint in most cases, which indicates that more rate is required in order to achieve lower coding complexity. Moreover, a coding for computing problem with side information is analogously studied. We also develop graph characterizations, which simplifies the computation of the achievable rate region.
title Distributed Approximate Computing with Constant Locality
topic Information Theory
url https://arxiv.org/abs/2312.04141