用快实验+慢实验结合，快速优化长期效果。
Experimenting, Fast and Slow: Bayesian Optimization of Long-term Outcomes with Online Experiments
- 融合短期快实验与长期慢实验，用贝叶斯方法迭代优化。
- 在短时间内完成对复杂系统策略的长期效果优化。
- 适合需要快速决策但关注长期影响的在线系统优化场景。
互联网系统中的在线实验（即A/B测试）广泛用于系统调优，例如优化推荐系统的排序策略和自适应流媒体控制器。决策者通常希望优化系统变更的长期处理效果，但由于处理效果随时间非平稳，短期测量可能具有误导性，因此需要长时间运行实验。传统顺序实验策略需多轮迭代，耗时过长。本文提出一种新方法：将短期快实验（如仅运行数小时或数天的偏差实验）和/或离线代理（如脱策略评估）与长期慢实验相结合，实现大动作空间内高效、贝叶斯序贯优化，在极短时间内完成长期效果优化。
原文摘要 · Abstract (English)
Online experiments in internet systems, also known as A/B tests, are used for a wide range of system tuning problems, such as optimizing recommender system ranking policies and learning adaptive streaming controllers. Decision-makers generally wish to optimize for long-term treatment effects of the system changes, which often requires running experiments for a long time as short-term measurements can be misleading due to non-stationarity in treatment effects over time. The sequential experimentation strategies--which typically involve several iterations--can be prohibitively long in such cases. We describe a novel approach that combines fast experiments (e.g., biased experiments run only for a few hours or days) and/or offline proxies (e.g., off-policy evaluation) with long-running, slow experiments to perform sequential, Bayesian optimization over large action spaces in a short amount of time.
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。