测试时增强让置信预测集更小，提升实用性
Test-time augmentation improves efficiency in conformal prediction
- 推理时引入测试时增强，利用归纳偏置缩小预测集
- 平均减少10%-14%的预测集合大小，且无需重训练模型
- 适合需要高效、可靠预测结果的场景，如医疗诊断
置信分类器生成一个预测类别集合，并提供概率保证该集合包含真实类别。然而，置信分类器常产生信息量不足的大集合。本文表明，测试时增强（TTA）——一种在推理阶段引入归纳偏置的技术——可有效缩小置信分类器生成的集合。该方法灵活、计算高效，可与任意置信评分结合，无需模型重训练，平均使预测集合缩小10%-14%。我们在三个数据集、三种模型、两种标准置信评分方法、不同置信强度及多种分布偏移下进行评估，验证了测试时增强在何时何地对置信框架有实际价值。
原文摘要 · Abstract (English)
A conformal classifier produces a set of predicted classes and provides a probabilistic guarantee that the set includes the true class. Unfortunately, it is often the case that conformal classifiers produce uninformatively large sets. In this work, we show that test-time augmentation (TTA)--a technique that introduces inductive biases during inference--reduces the size of the sets produced by conformal classifiers. Our approach is flexible, computationally efficient, and effective. It can be combined with any conformal score, requires no model retraining, and reduces prediction set sizes by 10%-14% on average. We conduct an evaluation of the approach spanning three datasets, three models, two established conformal scoring methods, different guarantee strengths, and several distribution shifts to show when and why test-time augmentation is a useful addition to the conformal pipeline.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。