arXiv：cs.LG· Samuel Daulton, David Eriksson, Maximilian Balandat, Eytan Bakshy·· 4 小时前AI 评分34
BONSAI：兼顾自然简洁与可解释性的贝叶斯优化方法
BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability
AI 导读
研究者提出 BONSAI，一种默认值感知的贝叶斯优化策略，可在保留默认配置的前提下剪除低影响偏离，并显式控制采集值的损失。理论分析表明，在适当条件下 BONSAI 保留了 vanilla GP-UCB 的无遗憾性质。实测中其候选生成开销平均仅为标准贝叶斯优化的 1.5 倍，而此前稀疏贝叶斯优化方法为 7-34 倍。
正文
Abstract:Bayesian optimization (BO) is a popular technique for sample-efficient optimization of black-box functions. In many applications, the parameters being tuned come with a carefully engineered default configuration, and practitioners only want to deviate from this default when necessary. Standard BO, however, does not aim to minimize deviation from the default and, in practice, often pushes weakly relevant parameters to the boundary of the search space. This makes it difficult to distinguish between important and spurious changes and increases the burden of vetting recommendations when the optimization objective omits relevant operational considerations. We introduce BONSAI, a default-aware BO policy that prunes low-impact deviations from a default configuration while explicitly controlling the loss in acquisition value. BONSAI is compatible with a variety of acquisition functions, including expected improvement and upper confidence bound (GP-UCB). We theoretically bound the regret incurred by BONSAI, showing that, under appropriate conditions, it retains the no-regret property of vanilla GP-UCB and removes irrelevant changes. Across many real-world applications, we empirically find that BONSAI substantially reduces the number of non-default parameters in recommended configurations while maintaining competitive optimization performance with little effect on wall time. Its candidate-generation cost averages only $1.5\times$ that of standard BO, compared with $7$-$34\times$ for prior sparse-BO methods.
| Comments: | 32 pages |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML) |
| Cite as: | arXiv:2602.07144 [cs.LG] |
| (or arXiv:2602.07144v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2602.07144 arXiv-issued DOI via DataCite |
|
| Journal reference: | NeurIPS 2026 |
Submission history
From: Samuel Daulton [view email]
[v1]
Fri, 6 Feb 2026 19:40:26 UTC (1,102 KB)
[v2]
Fri, 8 May 2026 22:04:33 UTC (1,116 KB)
[v3]
Wed, 7 Oct 2026 15:39:41 UTC (1,167 KB)
来源：arXiv：cs.LG · arxiv.org