提出信息增益评估虚拟染色模型预测的后验分布准确性
A Proper Scoring Rule for Virtual Staining
- 用信息增益作为逐细胞评估框架，直接衡量后验分布质量
- 在高通量筛选数据集上发现传统指标忽略的模型性能差异
- 理论严谨且可解释，适合跨模型、跨特征比较
生成式虚拟染色（VS）模型可用于高通量筛选（HTS），为每个输入和细胞提供可能生物特征值的估计后验分布。然而，评估时真实后验不可知。现有评估方法仅检验数据集上边缘分布的准确性，而非预测后验。本文引入信息增益（IG）作为细胞级评估框架，可直接评估预测后验。IG 是严格合适的评分规则，具有坚实的理论基础，支持可解释性，并能跨模型和特征比较。我们在一个广泛的HTS数据集上，使用IG及其他指标评估了基于扩散模型和GAN的模型，结果表明IG能揭示其他指标无法察觉的显著性能差异。
原文摘要 · Abstract (English)
Generative virtual staining (VS) models for high-throughput screening (HTS) can provide an estimated posterior distribution of possible biological feature values for each input and cell. However, when evaluating a VS model, the true posterior is unavailable. Existing evaluation protocols only check the accuracy of the marginal distribution over the dataset rather than the predicted posteriors. We introduce information gain (IG) as a cell-wise evaluation framework that enables direct assessment of predicted posteriors. IG is a strictly proper scoring rule and comes with a sound theoretical motivation allowing for interpretability, and for comparing results across models and features. We evaluate diffusion- and GAN-based models on an extensive HTS dataset using IG and other metrics and show that IG can reveal substantial performance differences other metrics cannot.
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。