In the Metaverse, real-time, concurrent services such as virtual classrooms and immersive gaming require local graphic rendering to maintain low latency. However, the limited processing power and battery capacity of user devices make it challenging to balance Quality of Experience (QoE) and terminal energy consumption. In this paper, we investigate a multi-objective optimization problem (MOP) regarding power control and rendering capacity allocation by formulating it as a multi-objective optimization problem. This problem aims to minimize energy consumption while maximizing Meta-Immersion (MI), a metric that integrates objective network performance with subjective user perception. To solve this problem, we propose a Multi-Objective Multi-Agent Evolutionary Reinforcement Learning with User-Object-Attention (M2ERL-UOA) algorithm. The algorithm employs a prediction-driven evolutionary learning mechanism for multi-agents, coupled with optimized rendering capacity decisions for virtual objects. The algorithm can yield a superior Pareto front that attains the Nash equilibrium. Simulation results demonstrate that the proposed algorithm can generate Pareto fronts, effectively adapts to dynamic user preferences, and significantly reduces decision-making time compared to several benchmarks.

Emerging use cases within the realm of industrial automation have underscored the importance of predictable wireless networked control when wireless networks bridge both real-time (RT) and time-sensitive (TS) services concurrently. However, the interplay between the varying fading channels, stochastic arrival tasks, and queuing states makes it challenging to guarantee completely the jitter-bounded deterministic latency of TS services and the throughput of RT services. To address this issue, we develop a predictable radio resource scheduling scheme based on Lyapunov-guided proximal policy optimization (LyPPO) for maximizing the transmission rate of RT services while adhering to the jitter-bounded deterministic delay constraint of TS services. The stochastic network calculus (SNC) is innovatively used to deduce the delay violation probability (DVP) and delay-constrained arrival rate bounds for RT services, which guides LyPPO by transforming the invisible jitter-bounded deterministic delay constraints of TS services into visible adaptive bandwidth limits. The simulation results verify that compared to alternative scheduling strategies, the proposed scheme adeptly mitigates the challenges posed by system dynamics and inherent uncertainties and provides predictable performance, encompassing both the foreseeability of TS services latency and the tolerability of RT services latency. Furthermore, the scheme exhibits superior performance in terms of packet loss probability and resource utilization efficiency.