提出跨噪声模型的安全贝叶斯优化方法，提升真实系统调控可靠性
Safe Bayesian optimization across noise models via scenario programming
- 用情景规划构建噪声概率边界，适配异方差与重尾分布
- 在模拟机械臂控制中实现安全优化，验证算法有效性
- 适合对安全性要求高的工业控制系统调参场景
安全贝叶斯优化（BO）结合高斯过程，是调优安全关键现实系统控制策略的有效工具，因其样本效率和安全保证。然而，多数安全BO算法假设测量噪声为同方差子高斯分布，这一假设在许多实际应用中不成立。本文提出一种简单但严谨的方法，适用于多种噪声模型，包括同方差子高斯噪声与异方差重尾分布。通过情景规划方法提供测量噪声的高概率界，并将其整合进高概率置信区间，证明了所提安全BO算法的安全性与最优性。我们在合成案例及仿真实验中对Franka Emika机械臂控制器进行调优，验证了算法性能。
原文摘要 · Abstract (English)
Safe Bayesian optimization (BO) with Gaussian processes is an effective tool for tuning control policies in safety-critical real-world systems, specifically due to its sample efficiency and safety guarantees. However, most safe BO algorithms assume homoscedastic sub-Gaussian measurement noise, an assumption that does not hold in many relevant applications. In this article, we propose a straightforward yet rigorous approach for safe BO across noise models, including homoscedastic sub-Gaussian and heteroscedastic heavy-tailed distributions. We provide a high-probability bound on the measurement noise via the scenario approach, integrate these bounds into high probability confidence intervals, and prove safety and optimality for our proposed safe BO algorithm. We deploy our algorithm in synthetic examples and in tuning a controller for the Franka Emika manipulator in simulation.
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。