用贝叶斯先验优化机器学习预测的统计推断，提升准确性并保持严格保证。
FAB-PPI: Frequentist, Assisted by Bayes, Prediction-Powered Inference
- 融合贝叶斯先验知识改进预测质量评估，增强推断可靠性
- 在预测质量符合先验预期时，估计更准确、置信区间更紧
- 对异常预测自动退化为标准PPI，适合高维数据与弱信号场景
预测驱动推断（PPI）通过结合实验数据与机器学习预测，实现有效的统计推断。当有足够多高质量预测时，PPI可获得比传统方法更精确的估计和更窄的置信区间。本文提出一种新方法FAB-PPI——频率派、贝叶斯辅助的预测驱动推断，利用对预测质量的先验知识改进框架。当实际预测质量与先验一致时，该方法优于标准PPI，同时保持频率派的严格保证。使用重尾先验时，FAB-PPI在先验概率低的区域自适应回归至标准PPI。我们在真实与合成数据上验证了其优势。
原文摘要 · Abstract (English)
Prediction-powered inference (PPI) enables valid statistical inference by combining experimental data with machine learning predictions. When a sufficient number of high-quality predictions is available, PPI results in more accurate estimates and tighter confidence intervals than traditional methods. In this paper, we propose to inform the PPI framework with prior knowledge on the quality of the predictions. The resulting method, which we call frequentist, assisted by Bayes, PPI (FAB-PPI), improves over PPI when the observed prediction quality is likely under the prior, while maintaining its frequentist guarantees. Furthermore, when using heavy-tailed priors, FAB-PPI adaptively reverts to standard PPI in low prior probability regions. We demonstrate the benefits of FAB-PPI in real and synthetic examples.
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。