发表机构
Faculty of Engineering, Bar-Ilan University（巴伊兰大学工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本论文针对标签噪声、无监督域适应及隐私保护场景，开发深度学习系统的置信度校准方法，提升其在实际部署中的可靠性与安全性。
AI 中文摘要
在高风险应用中，可靠的置信度估计与预测本身同等重要。置信度校准确保预测概率能反映正确的可能性，这对深度学习模型的安全部署至关重要。然而，现有方法通常假设可获取干净的验证数据，但由于标签噪声和域偏移，这一条件往往不现实。本论文开发了在这些条件下提升校准性能的方法。首先，我们解决标签噪声下的校准问题：当标签不可靠时，标准方法会产生误导性的置信度估计，我们提出一个框架，利用估计的噪声模型，通过建模带噪标签与干净标签分布之间的关系来重构无噪置信度估计；我们将此方法扩展到共形预测（Conformal Prediction, CP），该方法能提供具有保证覆盖率的集合值预测，我们的噪声感知共形预测方法可在存在标签噪声的情况下估计干净的一致性分数，从而实现可靠的不确定性量化。接下来，我们研究无监督域适应中的校准问题：模型在带标签的源域上训练后，需适配到无标签的目标域，由于无法获取目标域的带标签数据，我们从源域性能和域差异中估计目标域准确率，从而实现无需目标域标签的校准。我们还考虑隐私保护场景，其中用户标签和模型输出必须受到保护，我们提出一个局部差分隐私共形预测框架，该框架在提供有效不确定性量化的同时，维持隐私保证，并在隐私、计算可行性和预测可靠性之间取得平衡。我们的结果搭建了校准理论与安全关键应用实际部署之间的桥梁，为可靠、隐私保护且抗噪声的神经网络预测做出了贡献。
英文摘要
In high-stakes applications, reliable confidence estimates are as important as the predictions themselves. Confidence calibration ensures that predicted probabilities reflect the likelihood of correctness, making it essential for safe deployment of deep learning models. However, existing methods typically assume access to clean validation data, which is often unrealistic due to label noise and domain shifts. This thesis develops methods for improving calibration under these conditions. First, we address calibration under label noise. Standard methods can produce misleading confidence estimates when labels are unreliable. We propose a framework that uses an estimated noise model to reconstruct noise-free confidence estimates by modeling the relationship between noisy and clean label distributions. We extend this approach to Conformal Prediction (CP), which provides set-valued predictions with guaranteed coverage. Our noise-aware CP method estimates clean conformity scores despite label noise, enabling reliable uncertainty quantification. Next, we study calibration in unsupervised domain adaptation, where a model trained on a labeled source domain is adapted to an unlabeled target domain. Since labeled target data are unavailable, we estimate target-domain accuracy from source performance and domain discrepancies, enabling calibration without target labels. We also consider privacy-preserving settings in which user labels and model outputs must remain protected. We propose a locally differentially private conformal prediction framework that provides valid uncertainty quantification while maintaining privacy guarantees and balancing privacy, computational feasibility, and prediction reliability. Our results bridge calibration theory and practical deployment in safety-critical applications, contributing to reliable, privacy-preserving, and noise-resilient neural network predictions.
