From Optimization to Learning: Dual-Approach Resource Allocation for Over-the-Air Edge Computing Under Execution Uncertainty

TL;DR

This paper introduces a dual framework combining optimization and deep reinforcement learning to improve resource allocation in over-the-air edge computing under uncertainty, enhancing efficiency and scalability.

Contribution

It develops a systematic transition from mathematical optimization to DRL for resource management in AirComp systems with execution uncertainty, addressing multi-user complexity.

Findings

Optimization algorithms provide performance guarantees for single-user cases.

DRL framework adapts to multi-user environments with interference and contention.

Numerical results show improved system performance with increased edge server density.

Abstract

The exponential proliferation of mobile devices and data-intensive applications in future wireless networks imposes substantial computational burdens on resource-constrained devices, thereby fostering the emergence of over-the-air computation (AirComp) as a transformative paradigm for edge intelligence.} To enhance the efficiency and scalability of AirComp systems, this paper proposes a comprehensive dual-approach framework that systematically transitions from traditional mathematical optimization to deep reinforcement learning (DRL) for resource allocation under execution uncertainty. Specifically, we establish a rigorous system model capturing execution uncertainty via Gamma-distributed computational workloads, resulting in challenging nonlinear optimization problems involving complex Gamma functions. For single-user scenarios, we design advanced block coordinate descent (BCD) and majorization-maximization (MM) algorithms, which yield semi-closed-form solutions with provable performance guarantees. However, conventional optimization approaches become computationally intractable in dynamic multi-user environments due to inter-user interference and resource contention. To this end, we introduce a Deep Q-Network (DQN)-based DRL framework capable of adaptively learning optimal policies through environment interaction. Our dual methodology effectively bridges analytical tractability with adaptive intelligence, leveraging optimization for foundational insight and learning for real-time adaptability. Extensive numerical results corroborate the performance gains achieved via increased edge server density and validate the superiority of our optimization-to-learning paradigm in next-generation AirComp systems.