对比两种分子设计优化方法，发现多目标算法更优。
A study of EHVI vs fixed scalarization for molecule design
- 用期望超体积改进法优化分子，比固定权重法更高效。
- 在三种任务中，新方法覆盖更多优质分子，收敛更快。
- 尤其适合数据少、需权衡多个目标的分子设计场景。
多目标贝叶斯优化（MOBO）为分子设计中的权衡问题提供了合理框架，但其相对于标量化的实证优势尚未充分探索。本文在严格控制条件下，将基于帕累托的MOBO策略——期望超体积改进（EHVI）——与固定权重标量基线（期望改进，EI）进行对比，使用相同的高斯过程代理模型和分子表示。在三个分子优化任务中，EHVI在帕累托前沿覆盖率、收敛速度和化学多样性方面均持续优于标量化的EI。尽管标量化方法有随机或自适应等灵活变体，但即使强性能的确定性实例在低数据环境下仍表现不佳。结果表明，在评估预算有限且权衡非平凡时，帕累托感知的获取策略具有实际优势。
原文摘要 · Abstract (English)
Multi-objective Bayesian optimization (MOBO) provides a principled framework for navigating trade-offs in molecular design. However, its empirical advantages over scalarized alternatives remain underexplored. We benchmark a simple Pareto-based MOBO strategy - Expected Hypervolume Improvement (EHVI) - against a simple fixed-weight scalarized baseline using Expected Improvement (EI), under a tightly controlled setup with identical Gaussian Process surrogates and molecular representations. Across three molecular optimization tasks, EHVI consistently outperforms scalarized EI in terms of Pareto front coverage, convergence speed, and chemical diversity. While scalarization encompasses flexible variants - including random or adaptive schemes - our results show that even strong deterministic instantiations can underperform in low-data regimes. These findings offer concrete evidence for the practical advantages of Pareto-aware acquisition in de novo molecular optimization, especially when evaluation budgets are limited and trade-offs are nontrivial.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。