Adaptive Scheduling for Machine Learning Tasks over Networks
This work addresses resource allocation challenges in distributed learning for applications like smart transportation and IoT, but it appears incremental as it focuses on linear regression with adaptive scheduling.
The paper tackles the problem of efficiently allocating shared communication resources for distributed linear regression tasks by exploiting data informativeness, resulting in algorithms that provide reliable performance guarantees.
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.