arXiv：cs.LG· Gabriel Diaz-Aylwin, Vignesh Gopakumar, Omkar Myatra, David Moulton, Lorenzo Zanisi, David S. Leslie, Henry B. Moss·· 4 小时前
状态保持约束下的贝叶斯优化：面向 Tokamak 偏滤器优化
Bayesian Optimisation under State-Preservation Constraints
AI 导读
研究提出一种状态保持约束下的贝叶斯优化方法，通过预先计算线性化约束响应在容差内的控制集合，将状态空间约束拉回设计空间，形成可高效采样的椭球候选区域，并在线更新线性响应映射、重建椭球。该方法在 Tokamak 偏滤器等离子体边界保持优化任务上完成了端到端验证。
正文
Abstract:In many engineering design problems, the objective and constraints depend on the state: the solution of a PDE determined by the design parameters. We consider improving a design while holding selected state observables near trusted values, which we call state preservation constraints. Constrained Bayesian optimisation handles these with a learnt feasibility model, but struggles with this problem's highly anisotropic feasible set. Our central idea is to pre-compute the set of controls whose linearised constraint response stays within tolerance, thereby pulling back the state-space constraint into design space. This linearisation defines an ellipsoid from which we can efficiently draw a large number of well-spread candidates. The underlying linear response map is refined online, and the ellipsoid is rebuilt accordingly. We demonstrate the method end-to-end on our key application - Tokamak divertor optimisation under plasma-boundary preservation.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.12150 [cs.LG] |
| (or arXiv:2610.12150v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.12150 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Henry Moss [view email]
[v1]
Thu, 8 Oct 2026 15:34:26 UTC (4,480 KB)
来源：arXiv：cs.LG · arxiv.org