🤖 AI Summary
Deep learning–based resource allocation requires frequent retraining to accommodate dynamic user utility demands (e.g., time-varying data rate constraints), hindering real-time adaptability and scalability.
Method: This paper proposes ALCOR, an online adaptive resource allocator integrating deep neural networks, time-sharing optimization, and iterative user activation control. It introduces an expectation-satisfaction mechanism coordinated with unconstrained resource allocation (URA) and provides theoretical convergence guarantees.
Contribution/Results: ALCOR is the first framework enabling on-the-fly utility demand adjustment without retraining. It supports both centralized and distributed deployment while maintaining computational efficiency. Experiments demonstrate that ALCOR significantly outperforms static baseline policies under dynamic utility requirements—achieving superior utility satisfaction, real-time responsiveness, and generalization across diverse traffic patterns and network topologies.
📝 Abstract
Deep learning (DL) based resource allocation (RA) has recently gained significant attention due to its performance efficiency. However, most related studies assume an ideal case where the number of users and their utility demands, e.g., data rate constraints, are fixed, and the designed DL-based RA scheme exploits a policy trained only for these fixed parameters. Consequently, computationally complex policy retraining is required whenever these parameters change. In this paper, we introduce a DL-based resource allocator (ALCOR) that allows users to adjust their utility demands freely, such as based on their application layer requirements. ALCOR employs deep neural networks (DNNs) as the policy in a time-sharing problem. The underlying optimization algorithm iteratively optimizes the on-off status of users to satisfy their utility demands in expectation. The policy performs unconstrained RA (URA)--RA without considering user utility demands--among active users to maximize the sum utility (SU) at each time instant. Depending on the chosen URA scheme, ALCOR can perform RA in either a centralized or distributed scenario. Derived convergence analyses provide guarantees for ALCOR's convergence, and numerical experiments corroborate its effectiveness.
