---
title: "成本敏感共形预测与人机协同弃权：不平衡高风险决策支持多领域基准研究"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.27143v1"
permalink: "https://feed.astralor.com/items/itm_29df1c364f52b627349458d4"
published: "2026-07-29T17:15:33.000Z"
score: 80
channel: "ai"
tags: ["Mondrian共形预测","不平衡分类","人机协同","共形预测","成本控制弃权","高风险决策"]
---

# 成本敏感共形预测与人机协同弃权：不平衡高风险决策支持多领域基准研究

## AI 点评

> 这篇论文是部署高风险AI系统者必读。它直接揭示了标准共形预测在真实不平衡场景中的致命缺陷，并给出了经过大规模验证的解决方案。对信用评分、欺诈检测、医疗诊断领域的产品经理和工程师来说，Mondrian CP加成本控制弃机的组合几乎是现成的实用工具。

## 速览

高风险决策系统（如信用评分、欺诈检测、医疗诊断和工业安全）面临严重的类别不平衡和不对称错误成本问题。标准边际共形预测（CP）虽能保证整体覆盖，但对稀有且成本高昂的少数类覆盖极差，最低仅0.5%。本研究通过15个真实不平衡表格数据集、7种分类模型、3种概率校准技术及3150次实验，证明Mondrian CP可将少数类平均覆盖率提升61.7个百分点（p<1e-80）。结合成本控制弃权机制，在合理人工审核预算下显著降低预期决策成本。论文还量化了不同数据集的盈亏平衡阈值，为部署无分布假设、成本感知的不确定性量化提供了实用指导。

## 关键点

- 标准边际共形预测对不平衡高风险决策中的少数类覆盖极差，最低仅0.5%。
- Mondrian CP在15个数据集上将少数类平均覆盖率提升61.7个百分点。
- 结合成本控制弃权机制，在人机协同场景下显著降低预期决策成本。
- 论文提供了数据相关的盈亏平衡阈值，指导实际部署。

## 全文

High-stakes decision systems in credit scoring, fraud detection, healthcare, and industrial safety require reliable uncertainty quantification under severe class imbalance and asymmetric error costs. Standard marginal conformal prediction (CP) provides valid overall coverage guarantees; however, we show that it severely under-covers rare, costly minority classes, with minority-class coverage dropping to as low as 0.5% on certain datasets. To characterize and address this limitation, we conduct a comprehensive benchmark comparing marginal CP, class-conditional (Mondrian) CP, and cost-controlled abstention mechanisms across 15 real-world imbalanced tabular datasets, 7 classification models, 3 probability calibration techniques, and 10 random seeds, resulting in 3,150 experimental runs. Our results show that Mondrian CP restores valid minority-class coverage, achieving an average minority-coverage improvement of 61.7 percentage points over marginal CP (p < 1e-80). Furthermore, combining Mondrian CP with cost-controlled abstention significantly reduces expected decision cost compared with standard decision boundaries, confidence-based rejectors, and risk-controlled rejectors under realistic human review budgets. We further quantify dataset-specific break-even thresholds at which deferring ambiguous instances to human experts becomes cost-effective. These findings provide practical guidance for deploying distribution-free, cost-aware uncertainty quantification in high-stakes decision support systems.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_29df1c364f52b627349458d4