发表机构
University of Mannheim（曼海姆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出RelICL方法，利用表格基础模型通过模式图逐步传播信息，解决深度特征合成中的特征爆炸与交互盲区问题，在RelBench任务上与最强DFS方法性能相当。
AI 中文摘要
表格基础模型在无需任何训练的情况下，在单表任务上达到了最先进的性能。近期研究表明，它们同样适用于通过深度特征合成（DFS）进行的关系学习，该方法通过添加其他表列的聚合值作为特征，将关系模式展平为单个表格。这种方法颇具吸引力，因为它直接受益于底层表格基础模型的改进或定制化。在本文中，我们识别出DFS的两个关键问题：特征爆炸和交互盲区。第一个问题源于DFS特征数量随模式复杂度增加而快速增长，限制了可扩展性和性能。第二个问题源于列级聚合未考虑特征交互，同样限制了性能。我们提出并探索了一种名为RelICL的替代方法，它保留了DFS的优势，同时缓解了这两个问题。RelICL的核心在于，通过模式图逐步传播和融合信息，并利用最终用于预测的同一表格基础模型来完成这一过程。在我们使用RelBench任务的实验研究中，RelICL与基于深度特征合成的最强方法表现相当。
英文摘要
Tabular foundation models achieve state-of-the-art performance on single-table tasks without any training. Recent work suggests that they are also well-suited for relational learning via deep feature synthesis (DFS), which flattens a relational schema into a single table by adding aggregates of the other tables' columns as features. This approach is appealing because it directly benefits from improvements to or customization of the underlying tabular foundation model. In this paper, we identify two key problems with DFS: feature explosion and interaction blindness. The first problem arises because the number of DFS features grows quickly as the schema becomes more complex, limiting scalability and performance. The second problem arises because column-wise aggregates do not account for feature interactions, limiting performance. We propose and explore an alternative method termed RelICL, which keeps the benefits of DFS but alleviates these two problems. At its heart, RelICL propagates and fuses information step by step through the schema graph, using the same tabular foundation model that is eventually used for prediction to do so. In our experimental study using RelBench tasks, RelICL was on par with the strongest approach based on deep feature synthesis.
