E-Tree Learning: A Novel Decentralized Model Learning Framework for Edge AI
Lei Yang, Yanyan Lu, Jiannong Cao, Jiaming Huang, Mingjin Zhang
Abstract
Traditionally, Artificial Intelligence (AI) models are trained on the central cloud with data collected from end devices. This leads to high communication cost, long response time, and privacy concerns. Recently Edge-empowered AI, namely, Edge AI, has been proposed to support AI model learning and deployment at the network edge closer to the data sources. Existing research, including federated learning adopts a centralized architecture for model learning, where a central server aggregates the model updates from the clients/workers. The centralized architecture has drawbacks, such as performance bottleneck, poor scalability, and single point of failure. In this article, we propose a novel decentralized model learning approach, namely, E-Tree, which makes use of a well-designed tree structure imposed on the edge devices. The tree structure and the locations and orders of the aggregation on the tree are optimally designed to improve the training convergency and model accuracy. In particular, we design an efficient device clustering algorithm, named by K-Means and average accuracy, for E-Tree by taking into account the data distribution on the devices as well as the network distance. Evaluation results show that E-Tree significantly outperforms the benchmark approaches, such as federated learning and gossip learning under nonindependently and identically distributed (Non-i.i.d.) data in terms of model accuracy and convergency.