Artificial Intelligence Techniques for Deep Learning with Optimization-Based Task Scheduling and Computing Resource Allocation for VR Video Services in Advanced 6G Networks: Trends and Challenges

Abstract

The emergence of advanced 6G networks is expected to revolutionize immersive applications such as Virtual Reality (VR) video services, which demand ultra-low latency, high bandwidth, and efficient computational resource allocation. However, the massive data generation and stringent Quality of Experience (QoE) requirements pose significant challenges for task scheduling and computing resource allocation. Traditional optimization and heuristic-based approaches are insufficient to handle the dynamic, heterogeneous, and large-scale nature of 6G-enabled VR environments. This paper presents a comprehensive review of Artificial Intelligence (AI)-driven techniques, particularly deep learning combined with optimization methods, for task scheduling and resource allocation in VR video services. Deep Reinforcement Learning (DRL), Graph Neural Networks (GNNs), and hybrid optimization frameworks have emerged as promising solutions for addressing complex scheduling and resource allocation problems. Recent studies demonstrate that DRL-based approaches significantly improve latency, throughput, and resource utilization in dynamic 6G networks.mThe review analyses recent advancements (2020–2023), focusing on joint optimization frameworks, edge-cloud collaboration, and AI-driven scheduling mechanisms. It also highlights emerging trends such as knowledge-driven deep learning, federated learning, and intelligent edge computing. Furthermore, key challenges including computational complexity, scalability, real-time adaptability, and data privacy are discussed. The study concludes that hybrid AI models integrating deep learning with optimization techniques provide the most effective solutions for next-generation VR services in 6G networks. Future research should focus on lightweight, scalable, and energy-efficient AI frameworks capable of real-time deployment.