Surrogate-Powered Inference: Regularization and Adaptivity

2025-12-25

AI Summary

Under costly verification, scarce ground-truth labels and noisy proxy labels introduce estimation bias. To address this, we propose a unified manifold inference framework that jointly leverages ground-truth and proxy labels. Our method introduces a novel three-tier progressive SPI architecture—Base-SPI, SPI+, and SPI++—the first to jointly model adaptive multi-wave annotation and proxy-label regularization, thereby breaking the efficiency–bias trade-off bottleneck under constrained verification budgets. The framework integrates data augmentation, ℓ₂-regularized regression, and active learning–driven adaptive sampling, supported by asymptotic theory and Monte Carlo simulations. Experiments demonstrate substantial reductions in estimation error, improved statistical power for risk factor identification, and performance approaching that of fully ground-truth–labeled models—even with limited verification resources—enhancing model reliability and research reproducibility.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

High-quality labeled data are essential for reliable statistical inference, but are often limited by validation costs. While surrogate labels provide cost-effective alternatives, their noise can introduce non-negligible bias. To address this challenge, we propose the surrogate-powered inference (SPI) toolbox, a unified framework that leverages both the validity of high-quality labels and the abundance of surrogates to enable reliable statistical inference. SPI comprises three progressively enhanced versions. Base-SPI integrates validated labels and surrogates through augmentation to improve estimation efficiency. SPI+ incorporates regularized regression to safely handle multiple surrogates, preventing performance degradation due to error accumulation. SPI++ further optimizes efficiency under limited validation budgets through an adaptive, multiwave labeling procedure that prioritizes informative subjects for labeling. Compared to traditional methods, SPI substantially reduces the estimation error and increases the power in risk factor identification. These results demonstrate the value of SPI in improving the reproducibility. Theoretical guarantees and extensive simulation studies further illustrate the properties of our approach.

Problem

• Addresses bias from noisy surrogate labels in statistical inference
• Proposes a unified framework for integrating high-quality and surrogate labels
• Enhances estimation efficiency and power in risk factor identification

Innovation

• SPI integrates validated labels and surrogate labels
• SPI+ uses regularized regression for multiple surrogates
• SPI++ employs adaptive multiwave labeling for efficiency