让神经网络的不确定性预测更准，训练时直接对齐误差与预测置信度。
CLUE: Neural Networks Calibration via Learning Uncertainty-Error alignment
- 训练时显式对齐预测不确定性和实际误差，提升校准效果。
- 在视觉、语言等多任务上表现优于现有方法，且计算开销小。
- 完全可微、无需后处理，适合各类模型和场景部署。
可靠的不确定性估计对神经网络在真实场景中的应用至关重要。现有校准方法多依赖后处理或粗粒度分箱，存在可扩展性差、不可微、跨领域泛化能力弱等问题。本文提出CLUE（Calibration via Learning Uncertainty-Error Alignment），一种基于不确定性与误差对齐原则的新方法：良好校准的模型应使预测不确定性匹配其经验损失。CLUE采用新型损失函数，联合优化预测性能与校准性，以不确定性与损失的统计量作为代理信号。该方法完全可微、领域无关，兼容标准训练流程。在视觉、回归及语言建模任务上，包括分布外和域偏移场景的大量实验表明，CLUE在保持竞争力预测性能的同时，显著优于现有先进方法，且无显著计算负担。
原文摘要 · Abstract (English)
Reliable uncertainty estimation is critical for deploying neural networks (NNs) in real-world applications. While existing calibration techniques often rely on post-hoc adjustments or coarse-grained binning methods, they remain limited in scalability, differentiability, and generalization across domains. In this work, we introduce CLUE (Calibration via Learning Uncertainty-Error Alignment), a novel approach that explicitly aligns predicted uncertainty with observed error during training, grounded in the principle that well-calibrated models should produce uncertainty estimates that match their empirical loss. CLUE adopts a novel loss function that jointly optimizes predictive performance and calibration, using summary statistics of uncertainty and loss as proxies. The proposed method is fully differentiable, domain-agnostic, and compatible with standard training pipelines. Through extensive experiments on vision, regression, and language modeling tasks, including out-of-distribution and domain-shift scenarios, we demonstrate that CLUE achieves superior calibration quality and competitive predictive performance with respect to state-of-the-art approaches without imposing significant computational overhead.
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。