Fairness-Oracular MARL with Competitor-Aware Signals for Collaborative Inference

Hansong Zhou, Xiaonan Zhang

Keywords: Collaborative inference, Mean-field approximation, Fairness, Reinforcement learning

Abstract: Collaborative inference (CI) in NextG networks enables battery-powered devices to collaborate with nearby edges on deep learning inference. The fairness issue in a multi-device, multi-edge (M2M) CI system remains underexplored. Mean-field multi-agent reinforcement learning (MFRL) is a promising solution due to its low complexity and adaptability to system dynamics. However, the mobile nature of M2M CI systems hinders their effectiveness, as it breaks the premise of stable mean-field statistics. We propose FOCI (Fairness-Oriented Collaborative Inference), an RL-based method with two components: (i) an oracle-shaping reward for approaching max-min fairness and (ii) a competitor-aware observation augmentation for stabilizing device behaviors. We provide a convergence guarantee with bounded estimation errors. According to the results from real-world device mobility traces, FOCI demonstrates the best performance across multiple metrics and tightens the tails. It reduces worst-case latency by up to 56\% and worst-case energy by 46\% compared to baselines, while halving the switch cost and preserving competitive QoS.

Submission Number: 72