Federated Learning for Collaborative Inference Systems: The Case of Early Exit Networks
This work addresses performance gaps in IoT and edge computing by improving model training for collaborative inference, though it is incremental as it builds on existing federated learning and CIS methods.
The paper tackles the problem of training models for Cooperative Inference Systems (CISs) with heterogeneous client serving rates, proposing a novel federated learning approach that provides theoretical guarantees and outperforms state-of-the-art algorithms, achieving up to 15% higher accuracy in imbalanced scenarios.
As Internet of Things (IoT) technology advances, end devices like sensors and smartphones are progressively equipped with AI models tailored to their local memory and computational constraints. Local inference reduces communication costs and latency; however, these smaller models typically underperform compared to more sophisticated models deployed on edge servers or in the cloud. Cooperative Inference Systems (CISs) address this performance trade-off by enabling smaller devices to offload part of their inference tasks to more capable devices. These systems often deploy hierarchical models that share numerous parameters, exemplified by Deep Neural Networks (DNNs) that utilize strategies like early exits or ordered dropout. In such instances, Federated Learning (FL) may be employed to jointly train the models within a CIS. Yet, traditional training methods have overlooked the operational dynamics of CISs during inference, particularly the potential high heterogeneity in serving rates across clients. To address this gap, we propose a novel FL approach designed explicitly for use in CISs that accounts for these variations in serving rates. Our framework not only offers rigorous theoretical guarantees, but also surpasses state-of-the-art (SOTA) training algorithms for CISs, especially in scenarios where inference request rates or data availability are uneven among clients.