Offline Learning-Based Multi-User Delay-Constrained Scheduling

Effective multi-user delay-constrained scheduling is crucial in various real-world applications, such as instant messaging, live streaming, and data center management. In these scenarios, schedulers must make real-time decisions to satisfy both delay and resource constraints without prior knowledge of system dynamics, which are often time-varying and challenging to estimate. Current learning-based methods typically require real-time interaction with actual systems during the training stage, which can be difficult or impractical as it may degrade system performance and incur significant service costs. To address these challenges, we propose Scheduling by Offline Learning with Actor Rectification (SOLAR), an offline reinforcement learning-based algorithm designed to learn efficient scheduling policies purely from offline data. SOLAR learns policies exclusively from available datasets, eliminating the need for real-time interactions with the system. Experimental results demonstrate that SOLAR is resilient to various system dynamics, including partially observable environments, and delivers superior performance compared to existing methods.