为离线优化生成的候选设计提供可信的性能保证
Conformal Candidate Certification for Offline Model-Based Optimization
- 用校准后的下界为每个候选设计打分，确保可靠性
- 在90%置信水平下实现99.0%的实际覆盖率，提升超16倍
- 适合对安全性和可靠性要求高的工业设计场景
离线模型基于优化（MBO）通过优化一个固定历史数据集训练的代理模型来生成候选设计。由于这些候选设计故意位于分布外，代理模型的排序在优化最激进处最不可靠。现有方法无法为每个候选提供满足目标阈值的统计证书。本文提出后处理封装方法「共形候选认证」（CCC），为每个候选附加一个校准的一侧下界，并仅推进那些下界超过目标值的候选。我们证明熵正则化代理最大化会诱导一个吉布斯倾斜提议，因此同一代理可直接提供加权共形预测的重要性权重，无需额外密度比估计步骤。在受控合成实验中，CCC 在名义 0.90 置信水平下实现了 0.990 的实际覆盖率达 16.7%，而忽略协变量偏移的标准共形预测覆盖率暴跌至 0.416。
原文摘要 · Abstract (English)
Offline model-based optimization (MBO) proposes candidates by optimizing a surrogate trained on a fixed historical dataset. Because candidates are deliberately out-of-distribution, surrogate rankings are least reliable exactly where the optimizer is most aggressive, yet existing methods provide no per-candidate statistical certificate that a design meets a target threshold. We propose \emph{Conformal Candidate Certification} (CCC), a post-hoc wrapper that attaches a calibrated one-sided lower bound to each candidate and advances only those whose bound exceeds the target. We show that entropy-regularized surrogate maximization induces a Gibbs-tilted proposal, so the same surrogate supplies importance weights for weighted conformal prediction without a separate density-ratio estimation step. In a controlled synthetic study, CCC certifies $16.7\%$ of an aggressive proposal pool with empirical coverage 0.990 at nominal 0.90, while standard conformal prediction ignoring the covariate shift collapses to 0.416 coverage.
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。