User Experience Perception-Oriented Network Optimization Method Based on Reinforcement Learning

This paper proposes a novel approach for network optimization by utilizing reinforcement learning, aimed at enhancing user experience perception. The proposed method constructs a comprehensive user perception system, mapping network indicators to user experience and aggregating perceptions across multiple dimensions. By incorporating a reward mechanism for training reinforcement learning agents, the method generates network optimization strategies that prioritize user perception. This approach significantly improves the real experience of network users, offering a promising direction for future research and practical applications in network optimization.