📄 中文摘要
安全强化学习作为一种新兴的研究领域，旨在确保智能体在学习过程中遵循安全约束，避免潜在的危险和不良后果。该领域的研究主要集中在几种关键方法上，包括基于模型的安全策略、约束优化技术以及安全探索机制。这些方法不仅在理论上具有重要意义，还在实际应用中展现了广泛的前景，如自动驾驶、机器人控制和金融决策等。通过对现有文献的系统性回顾，识别出当前面临的挑战和未来的发展方向，强调了安全性在强化学习中的重要性和必要性。
安全强化学习的综述：方法、理论与应用
📄 English Summary
A Review of Safe Reinforcement Learning: Methods, Theory and Applications
Safe reinforcement learning (SRL) has emerged as a critical area of research focused on ensuring that agents adhere to safety constraints during the learning process, thereby avoiding potential hazards and adverse outcomes. Key methods in this field include model-based safety policies, constraint optimization techniques, and safe exploration mechanisms. These approaches hold significant theoretical implications and demonstrate broad prospects in practical applications such as autonomous driving, robotic control, and financial decision-making. A systematic review of existing literature identifies current challenges and future directions, emphasizing the importance and necessity of safety in reinforcement learning.