发表机构
MOGAM Institute for Biomedical Research; CROID Research; aSSIST University（MOGAM生物医学研究所; CROID研究; aSSIST大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对标签偏移下共形预测失效问题，提出标签偏移调整的贝叶斯分数（LSA分数），通过后验预测倾斜恒等式修正贝叶斯分数，在分子性质预测中生成更短区间且覆盖相当。
AI 中文摘要
共形预测在可交换性假设下提供了无分布的量化不确定性。然而，这一假设在标签偏移情况下被违反，此时标签的边缘分布发生变化，而给定标签的输入条件分布保持稳定。在这种偏移下，标准共形程序不再维持其预期的覆盖行为。现有方法通过重要性加权来解决这一问题。它们将重新加权与基于残差的非一致性分数配对，后者忽略了预测不确定性，导致生成的区间具有均匀宽度。贝叶斯共形方法通过利用预测分布产生自适应区间，它们在源预测下评估一致性，这在标签偏移下与目标域不对齐。我们提出了标签偏移调整的贝叶斯分数（LSA分数），这是一种从后验预测倾斜恒等式导出的非一致性分数。该恒等式表明目标预测是源预测的重要性加权变换。我们利用它推导出对贝叶斯分数的直接修正。我们在受控标签偏移下对分子性质预测进行了方法评估。LSA分数始终比基于残差和基于源的贝叶斯分数产生更短的区间。目标域的覆盖范围保持可比。在更强的偏移下，由于基于伪标签的密度比估计，所有方法都会遭受一定的覆盖损失。LSA分数适用于任何具有可处理对数密度的源预测。我们使用贝叶斯岭回归实例化它，其中修正具有封闭形式。
英文摘要
Conformal prediction provides distribution-free uncertainty quantification under exchangeability. However, this assumption is violated by label shift, where the marginal distribution of labels changes while the conditional distribution of inputs given labels remains stable. Under such shifts, standard conformal procedures no longer maintain their intended coverage behavior. Existing approaches address this via importance weighting. They pair the reweighting with residual-based nonconformity scores that ignore predictive uncertainty. The resulting intervals have uniform width. Bayesian conformal methods produce adaptive intervals by leveraging predictive distributions. They evaluate conformity under the source predictive, which is misaligned with the target domain under label shift. We propose the \emph{Label-Shift-Adjusted Bayesian Score} (LSA score), a nonconformity score derived from a posterior predictive tilting identity. This identity shows that the target predictive is an importance-weighted transformation of the source predictive. We use it to derive a direct correction to the Bayesian score. We evaluate the method on molecular property prediction under controlled label shift. The LSA score consistently yields shorter intervals than residual-based and source-based Bayesian scores. Coverage in the target domain remains comparable. Under stronger shift, all methods incur some coverage loss due to pseudo-label-based density-ratio estimation. The LSA score is defined for any source predictive with a tractable log-density. We instantiate it with Bayesian Ridge Regression, where the correction admits a closed form.
Comments2nd Workshop on Epistemic Intelligence in Machine Learning (EIML@ICML 2026),
