LAVA：无监督潜在嵌入的可解释性
LAVA: Explainability for Unsupervised Latent Embeddings
AI总结：
LAVA是一种无监督嵌入可解释性方法，通过输入数据特征共变揭示局部嵌入结构，发现领域相关模式。
摘要：
无监督黑箱模型是科学发现的驱动力，但难以解释，因为其输出通常是多维嵌入而不是明确的目标。虽然监督学习的可解释性揭示了输入特征如何影响预测，其无监督 counterpart 应将输入特征与学习到的嵌入结构相关联。然而，将监督模型可解释性适应于无监督学习的改进要么提供单样本或数据集汇总解释，过于细粒度或简化，无法有意义地解释嵌入而不使用映射函数。为了弥合这一差距，我们提出了LAVA，一种事后模型无关的方法，通过原始输入数据中的特征共变来解释局部嵌入组织。LAVA的解释包含模块，捕捉输入特征相关性的局部子模式，这些子模式在嵌入中全球范围内重复出现。LAVA在所需粒度级别上提供稳定的解释，揭示了领域相关的模式，如图像的视觉部分或细胞过程中的疾病信号，这些模式现有方法未能发现。
英文摘要：
Unsupervised black-box models are drivers of scientific discovery, yet are difficult to interpret, as their output is often a multidimensional embedding rather than a well-defined target. While explainability for supervised learning uncovers how input features contribute to predictions, its unsupervised counterpart should relate input features to the structure of the learned embeddings. However, adaptations of supervised model explainability for unsupervised learning provide either single-sample or dataset-summary explanations, remaining too fine-grained or reductive to be meaningful, and cannot explain embeddings without mapping functions. To bridge this gap, we propose LAVA, a post-hoc model-agnostic method to explain local embedding organization through feature covariation in the original input data. LAVA explanations comprise modules, capturing local subpatterns of input feature correlation that reoccur globally across the embeddings. LAVA delivers stable explanations at a desired level of granularity, revealing domain-relevant patterns such as visual parts of images or disease signals in cellular processes, otherwise missed by existing methods.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
