迁移校准预测增强推断
Transfer Calibrated Prediction Powered Inference
- The Ohio State University（俄亥俄州立大学）
- Yale University（耶鲁大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出TC-PPI方法，通过交叉拟合和多种校准技术适应源域预测器到目标域，联合估计器Joint-TC-Cross-PPI++保证无偏且效率不低于现有方法，防止负迁移，并在实际应用中验证改进。
AI中文摘要：
预测增强推断（PPI）及其调优扩展（PPI++）通过将小规模金标准标注样本与大规模AI模型的预测相结合，来改进置信区间。其效率提升依赖于较低的残差方差，但若预测器是在不同源域上预训练的，这一条件可能不成立。我们提出迁移校准预测增强推断（TC-PPI），利用金标准样本通过交叉拟合将源域预测器适应到目标域。该方法支持多种适应技术，如稀疏线性校准、LoRA和微调。我们联合调优的交叉拟合估计器Joint-TC-Cross-PPI++保持无偏性，并且同时至少与经典推断、PPI和PPI++一样高效，从而防止负迁移。我们提供了高维均方误差界用于校准，并在多种实际应用中展示了相对于基线方法的经验改进。
英文摘要：
Prediction-powered inference (PPI) and its power-tuned extension (PPI++) improve confidence intervals by combining a small gold-standard labeled sample with a large AI model's predictions. Its efficiency gain relies on low residual variance, which may not hold if the predictor is pre-trained on a different source domain. We propose Transfer Calibrated Prediction-Powered Inference (TC-PPI), adapting the source-domain predictor to the target domain using gold-standard samples through cross-fitting. This approach supports various adaptation methods, such as sparse linear calibration, LoRA, and fine-tuning. Our jointly tuned cross-fit estimator, Joint-TC-Cross-PPI++, maintains unbiasedness and is simultaneously at least as efficient as classical inference, PPI, and PPI++, thereby protecting against negative transfer. We provide high-dimensional MSE bounds for calibration and show empirical improvements over baseline methods across various real-world applications.
