Adaptive Scheduling for Machine Learning Tasks over Networks

TL;DR

This paper introduces adaptive scheduling algorithms for distributed machine learning tasks over shared networks, optimizing resource allocation based on data informativeness to improve efficiency and performance guarantees.

Contribution

It presents novel algorithms for adaptive scheduling in distributed learning that leverage data informativeness, with proven performance guarantees.

Findings

Algorithms effectively allocate resources in distributed settings.

Adaptive scheduling improves learning efficiency.

Performance guarantees are established for the proposed methods.

Abstract

A key functionality of emerging connected autonomous systems such as smart transportation systems, smart cities, and the industrial Internet-of-Things, is the ability to process and learn from data collected at different physical locations. This is increasingly attracting attention under the terms of distributed learning and federated learning. However, in this setup data transfer takes place over communication resources that are shared among many users and tasks or subject to capacity constraints. This paper examines algorithms for efficiently allocating resources to linear regression tasks by exploiting the informativeness of the data. The algorithms developed enable adaptive scheduling of learning tasks with reliable performance guarantees.