arXiv：cs.LG· Michal Kmicikiewicz, Tommy Rochussen, Vincent Fortuin, Ewa Szczurek·· 4 小时前AI 评分34
MetaHeta：在测定异质性下用元学习做小样本生物活性预测
Few-Shot Bioactivity Prediction with Meta-Learning under Assay Heterogeneity
AI 导读
研究提出 MetaHeta 元学习框架，通过引入相关测定的辅助数据来应对测定异质性，以改善小样本生物活性预测。该框架结合对大规模辅助数据的线性注意力与对稀缺任务上下文的精确注意力，从而在不牺牲后者精度的前提下高效扩展前者。在 ChEMBL 和 BindingDB 的测定数据上，MetaHeta 提升了小样本预测表现及回顾性贝叶斯优化中的化合物优选效果。
正文
Abstract:Accurate bioactivity prediction is a central challenge in early-stage drug discovery, as individual assays often contain too few measurements to train reliable models independently. Meta-learning offers a principled approach to this few-shot setting, but assay heterogeneity may limit its effectiveness. Here, we test this hypothesis and show that meta-learning performance degrades as meta-training tasks become more heterogeneous. To address this, we introduce MetaHeta, a meta-learning framework that accounts for assay heterogeneity by conditioning predictions on auxiliary data from related assays, with relatedness defined flexibly from available assay information. The architecture of MetaHeta combines linear attention over large auxiliary datasets with exact attention over scarce task-specific context, enabling efficient scaling to the former without compromising exact attention over the latter. We demonstrate the benefits of our approach on assays from ChEMBL and BindingDB, improving few-shot bioactivity prediction and downstream compound prioritization in retrospective Bayesian optimization.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.07079 [cs.LG] |
| (or arXiv:2610.07079v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07079 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Michal Kmicikiewicz [view email]
[v1]
Mon, 5 Oct 2026 11:01:11 UTC (1,557 KB)
来源：arXiv：cs.LG · arxiv.org