研究人员介绍了评估机器学习模型中不确定性量化（UQ）的新方法。一种称为“决策对齐”的方法旨在确保UQ指标与下游决策效用有意义地相关，从而揭示了当前通用指标的缺陷。另一项进展侧重于多模态回归任务的高效UQ，将变分贝叶斯推理扩展到分位数回归和分类恢复等模型。此外，还提出了一种名为Ribbon的可扩展近似方法，该方法通过近似贝叶斯自举方法而不要求重复模型重新拟合，从而提供鲁棒的不确定性量化。 AI
影响 不确定性量化的进步可能带来更可靠、更值得信赖的人工智能系统，尤其是在关键决策应用中。
排序理由 多篇在arXiv上发表的研究论文介绍了机器学习中不确定性量化方面的新方法和评估。
AI 生成摘要 · Google Gemini · 来自 5 个来源。 我们如何撰写摘要 →
Uncertainty estimates in machine learning are typically evaluated using generic metrics such as the negative log-likelihood and expected calibration error, yet good performance on such metrics does not necessarily imply high utility in downstream decisions. We introduce decision-…
arXiv:2606.25188v1 Announce Type: new Abstract: Efficient uncertainty quantification (UQ) is essential for trustworthy large-scale learning. Existing UQ methods for regression tasks mainly operate under the assumption that the conditional label marginal satisfies single-peak para…
arXiv:2606.27269v1 Announce Type: new Abstract: Reliably quantifying predictive uncertainty is difficult for complex, high-dimensional, or misspecified models. Both fully Bayesian and bootstrap resampling methods provide principled uncertainty estimates but are often too expensiv…
arXiv:2606.26990v1 Announce Type: cross Abstract: Uncertainty estimates in machine learning are typically evaluated using generic metrics such as the negative log-likelihood and expected calibration error, yet good performance on such metrics does not necessarily imply high utili…
Reliably quantifying predictive uncertainty is difficult for complex, high-dimensional, or misspecified models. Both fully Bayesian and bootstrap resampling methods provide principled uncertainty estimates but are often too expensive for modern machine-learning models because the…