arXiv：cs.LG· Aime Bienfait Igiraneza, Christophe Fraser, Robert Hinch·· 5 小时前AI 评分27
Precise Quantifier（PQ）：一种实现窄预测区间与充分覆盖的贝叶斯聚合量化器
Estimating prevalence with precision and accuracy
AI 导读
研究者提出 Precise Quantifier（PQ），一种贝叶斯聚合量化器，用于估计数据集中类别分布并量化估计不确定性，可实现窄预测区间与充分覆盖。实验显示，随着底层分类器判别力提升及验证集与测试集规模比增大，PQ 的流行率估计比现有方法更精确，表明其能更有效利用验证信息量化不确定性。
正文
Abstract:Unlike classification, whose goal is to estimate the class of each data point, quantification (or prevalence estimation) aims to estimate the distribution of classes in a dataset. An important task in prevalence estimation is to quantify the uncertainty in prevalence estimates. In this paper, we introduce Precise Quantifier (PQ), a Bayesian aggregative quantifier that achieves narrow prediction intervals with sufficient coverage (i.e., sufficient proportion of intervals containing the true prevalence). We find that PQ produces more precise prevalence estimates than existing methods as the discriminative power of the underlying classifier increases and as the validation-to-test size ratio increases. These empirical results suggest that PQ uses validation information more effectively to quantify uncertainty in prevalence estimates than existing approaches.
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG) |
| Cite as: | arXiv:2507.06061 [stat.ML] |
| (or arXiv:2507.06061v2 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2507.06061 arXiv-issued DOI via DataCite |
Submission history
From: Aime Bienfait Igiraneza [view email]
[v1]
Tue, 8 Jul 2025 15:06:02 UTC (2,865 KB)
[v2]
Fri, 2 Oct 2026 11:49:10 UTC (3,449 KB)
来源：arXiv：cs.LG · arxiv.org