Robustness of Decentralised Learning to Nodes and Data Disruption

TL;DR

This paper investigates the robustness of decentralised learning systems against node and data disruptions, demonstrating that such systems can recover and maintain classification accuracy even after significant disruptions, depending on data distribution and network connectivity.

Contribution

It provides a comprehensive analysis of how decentralised learning withstands node disruptions, highlighting the importance of data persistence and network structure for robustness.

Findings

Decentralised learning remains robust with minimal data remaining.

Knowledge retention is possible even with isolated nodes.

Network connectivity influences recovery from disruptions.

Abstract

In the vibrant landscape of AI research, decentralised learning is gaining momentum. Decentralised learning allows individual nodes to keep data locally where they are generated and to share knowledge extracted from local data among themselves through an interactive process of collaborative refinement. This paradigm supports scenarios where data cannot leave local nodes due to privacy or sovereignty reasons or real-time constraints imposing proximity of models to locations where inference has to be carried out. The distributed nature of decentralised learning implies significant new research challenges with respect to centralised learning. Among them, in this paper, we focus on robustness issues. Specifically, we study the effect of nodes' disruption on the collective learning process. Assuming a given percentage of "central" nodes disappear from the network, we focus on different cases, characterised by (i) different distributions of data across nodes and (ii) different times when disruption occurs with respect to the start of the collaborative learning task. Through these configurations, we are able to show the non-trivial interplay between the properties of the network connecting nodes, the persistence of knowledge acquired collectively before disruption or lack thereof, and the effect of data availability pre- and post-disruption. Our results show that decentralised learning processes are remarkably robust to network disruption. As long as even minimum amounts of data remain available somewhere in the network, the learning process is able to recover from disruptions and achieve significant classification accuracy. This clearly varies depending on the remaining connectivity after disruption, but we show that even nodes that remain completely isolated can retain significant knowledge acquired before the disruption.