发表机构
Université Laval; Mila - Quebec Artificial Intelligence Institute（拉瓦尔大学; 米拉-魁北克人工智能研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对生化数据机器学习模型评估时因未考虑关系结构致数据泄露问题，提出基于关系生成过程的Refnd划分算法，利用HNSW快速计算近邻图，在抗菌肽数据集验证，性能评估更现实且适用多种数据集。
AI 中文摘要
在生化数据上训练的机器学习模型，通常使用未考虑关系结构的划分进行评估，导致信息泄露和性能估计过于乐观。现有划分方法缺乏理论基础且扩展性差。我们引入关系生成过程（RGP）来解释生化数据集中关系结构的产生，并提出Refnd划分算法，它利用分层可导航小世界（HNSW）在对数线性时间内计算的近邻图。在抗菌肽数据集上验证表明，Refnd划分产生的评估性能虽低但更现实。Refnd适用于RGP产生的任何数据集，可通过pip install refnd安装。
英文摘要
Machine learning models trained on biochemical data are routinely evaluated using splits that fail to account for relational structure, causing information leakage and over-optimistic performance estimates. Existing splitting methods lack theoretical grounding and scale at best quadratically. We introduce the Relational Generative Process (RGP), a mathematical formalization explaining why relational structure arises in biochemical datasets, and Refnd, a splitting algorithm that leverages a proximity graph computed in loglinear time using Hierarchical Navigable Small World (HNSW). We validate on an antimicrobial peptide dataset, showing that Refnd splits yield lower but more realistic evaluation performance than traditional splits. Refnd is applicable to any dataset arising from an RGP such as protein sequences and structures, small molecules, and nucleotide sequences, and is openly available as a Rust accelerated Python package: pip install refnd.
