诊断和修复分子设计中贝叶斯优化的常见问题
Diagnosing and fixing common problems in Bayesian optimization for molecule design
AI总结:
本文指出了分子设计中贝叶斯优化的三个常见陷阱(先验宽度不当、过度平滑、采集函数最大化不足),并证明修复后基本BO设置在PMO基准上取得最高性能,呼吁分子机器学习社区给予BO更多关注。
摘要:
贝叶斯优化(BO)是分子设计任务的一种有原则的方法。在本文中,我们解释了BO的三个可能导致较差经验性能的陷阱:不正确的先验宽度、过度平滑和不足的采集函数最大化。我们表明,在解决了这些问题后,即使是基本的BO设置也能在分子设计的PMO基准(Gao等人2022)上实现最高的整体性能。这些结果表明,BO可能会从分子机器学习社区的更多关注中受益。
英文摘要:
Bayesian optimization (BO) is a principled approach to molecular design tasks. In this paper we explain three pitfalls of BO which can cause poor empirical performance: an incorrect prior width, over-smoothing, and inadequate acquisition function maximization. We show that with these issues addressed, even a basic BO setup is able to achieve the highest overall performance on the PMO benchmark for molecule design (Gao et al 2022). These results suggest that BO may benefit from more attention in the machine learning for molecules community.
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。