Task-Aware Network Coding Over Butterfly Network

TL;DR

This paper introduces a task-aware network coding framework over butterfly networks, leveraging machine learning and PCA to transmit task-relevant data efficiently, with theoretical bounds and algorithms demonstrating improved performance.

Contribution

It formulates a novel task-driven network coding problem, derives bounds, and proposes ML algorithms for efficient transmission of task-relevant data in butterfly networks.

Findings

Lower bound for total loss function established

Necessary and sufficient conditions for optimality derived

ML algorithms effectively solve the task-aware coding problem

Abstract

Network coding allows distributed information sources such as sensors to efficiently compress and transmit data to distributed receivers across a bandwidth-limited network. Classical network coding is largely task-agnostic -- the coding schemes mainly aim to faithfully reconstruct data at the receivers, regardless of what ultimate task the received data is used for. In this paper, we analyze a new task-driven network coding problem, where distributed receivers pass transmitted data through machine learning (ML) tasks, which provides an opportunity to improve efficiency by transmitting salient task-relevant data representations. Specifically, we formulate a task-aware network coding problem over a butterfly network in real-coordinate space, where lossy analog compression through principal component analysis (PCA) can be applied. A lower bound for the total loss function for the formulated problem is given, and necessary and sufficient conditions for achieving this lower bound are also provided. We introduce ML algorithms to solve the problem in the general case, and our evaluation demonstrates the effectiveness of task-aware network coding.