Argus: Federated Non-convex Bilevel Learning over 6G Space-Air-Ground Integrated Network
This addresses the problem of inefficient optimization in infrastructureless, time-varying networks for applications like autonomous aerial vehicles, though it appears incremental as it adapts existing federated learning concepts to a specific network setting.
The paper tackles the challenge of decentralized federated bilevel learning in 6G space-air-ground integrated networks by proposing Argus, an asynchronous algorithm that avoids stragglers, with theoretical analysis and numerical experiments showing its effectiveness.
The space-air-ground integrated network (SAGIN) has recently emerged as a core element in the 6G networks. However, traditional centralized and synchronous optimization algorithms are unsuitable for SAGIN due to infrastructureless and time-varying environments. This paper aims to develop a novel Asynchronous algorithm a.k.a. Argus for tackling non-convex and non-smooth decentralized federated bilevel learning over SAGIN. The proposed algorithm allows networked agents (e.g. autonomous aerial vehicles) to tackle bilevel learning problems in time-varying networks asynchronously, thereby averting stragglers from impeding the overall training speed. We provide a theoretical analysis of the iteration complexity, communication complexity, and computational complexity of Argus. Its effectiveness is further demonstrated through numerical experiments.