English(EN) Split Conformal Prediction with Label-Shift-Adjusted Bayesian Scores

新的贝叶斯得分增强了标签偏移下的不确定性量化

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-14 04:00

研究人员开发了一种名为标签偏移调整贝叶斯得分（LSA score）的新方法，以在标签分布发生变化时改进一致性预测中的不确定性量化。标准的一致性预测方法在处理标签偏移时存在困难，会导致覆盖率不准确。LSA得分通过使用后验预测倾斜恒等式来调整贝叶斯得分，从而解决了这个问题，该恒等式考虑了标签分布的偏移。这种方法在分子性质预测方面显示出潜力，与现有方法相比，在具有可比覆盖率的情况下产生了更短的区间。 AI

影响在数据分布变化的场景下提高了AI预测的可靠性。

排序理由学术论文，详细介绍了机器学习中不确定性量化的一种新方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 23 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，详细介绍了机器学习中不确定性量化的一种新方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Hyeonsu Lee, Juyeon Kim, Erkhembayar Jadamba, Seungjin Choi, Hyunjin Shin · 2026-09-14 04:00

带有标签偏移调整的贝叶斯得分的分裂保形预测

arXiv:2609.12386v1 Announce Type: new Abstract: 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 distribut…

本文如何被排名

报道来源 [1]

带有标签偏移调整的贝叶斯得分的分裂保形预测

相关实体

相关话题