Practical sufficient conditions for convergence of distributed optimisation algorithms over communication networks with interference

TL;DR

This paper develops realistic network models with interference and delays, providing verifiable conditions under which distributed optimization algorithms reliably converge despite network imperfections.

Contribution

It introduces practical, locally verifiable network conditions ensuring convergence of distributed algorithms in interference-laden, delayed communication environments.

Findings

Convergence is guaranteed under conditions where AoI is in O(√n).

Distributed gradient descent converges even with diverging average AoI.

The model bridges practical network issues with abstract optimization theory.

Abstract

Information exchange over networks can be affected by various forms of delay. This causes challenges for using the network by a multi-agent system to solve a distributed optimisation problem. Distributed optimisation schemes, however, typically do not assume network models that are representative for real-world communication networks, since communication links are most of the time abstracted as lossless. Our objective is therefore to formulate a representative network model and provide practically verifiable network conditions that ensure convergence of distributed algorithms in the presence of interference and possibly unbounded delay. Our network is modelled by a sequence of directed-graphs, where to each network link we associate a process for the instantaneous signal-to-interference-plus-noise ratio. We then formulate practical conditions that can be verified locally and show that the age of information (AoI) associated with data communicated over the network is in $\mathcal{O}(\sqrt{n})$. Under these conditions we show that a penalty-based gradient descent algorithm can be used to solve a rich class of stochastic, constrained, distributed optimisation problems. The strength of our result lies in the bridge between practical verifiable network conditions and an abstract optimisation theory. We illustrate numerically that our algorithm converges in an extreme scenario where the average AoI diverges.