Initialisation and Network Effects in Decentralised Federated Learning
This work addresses efficiency and scalability for distributed neural network training in decentralized federated learning, offering incremental improvements in initialization methods.
The paper tackled the problem of training efficiency in decentralized federated learning by proposing an uncoordinated initialization strategy based on eigenvector centralities of the network topology, which led to radically improved training efficiency.
Fully decentralised federated learning enables collaborative training of individual machine learning models on a distributed network of communicating devices while keeping the training data localised on each node. This approach avoids central coordination, enhances data privacy and eliminates the risk of a single point of failure. Our research highlights that the effectiveness of decentralised federated learning is significantly influenced by the network topology of connected devices and the learning models' initial conditions. We propose a strategy for uncoordinated initialisation of the artificial neural networks based on the distribution of eigenvector centralities of the underlying communication network, leading to a radically improved training efficiency. Additionally, our study explores the scaling behaviour and the choice of environmental parameters under our proposed initialisation strategy. This work paves the way for more efficient and scalable artificial neural network training in a distributed and uncoordinated environment, offering a deeper understanding of the intertwining roles of network structure and learning dynamics.