用机器学习优化罕见事件估计，显著降低方差。
When Machine Learning Meets Importance Sampling: A More Efficient Rare Event Estimation Approach
- 基于稳态分布的边际似然比，避免路径依赖导致的方差爆炸
- 通过采样数据训练机器学习模型估算边际似然比
- 在排队系统中验证，性能优于经典重要性采样方法
针对通信网络中串联队列稳态下罕见事件概率估计问题，现有重要性采样方法因路径依赖的似然函数方差过大而效率低下。本文提出一种新方法，利用稳态分布的边际似然比，有效规避方差膨胀问题。同时设计基于重要性采样数据的机器学习算法，用于估计该边际似然比。数值实验表明，所提算法在多个测试场景下均显著优于经典重要性采样方法。
原文摘要 · Abstract (English)
Driven by applications in telecommunication networks, we explore the simulation task of estimating rare event probabilities for tandem queues in their steady state. Existing literature has recognized that importance sampling methods can be inefficient, due to the exploding variance of the path-dependent likelihood functions. To mitigate this, we introduce a new importance sampling approach that utilizes a marginal likelihood ratio on the stationary distribution, effectively avoiding the issue of excessive variance. In addition, we design a machine learning algorithm to estimate this marginal likelihood ratio using importance sampling data. Numerical experiments indicate that our algorithm outperforms the classic importance sampling methods.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。