Bayesian Optimisation under State-Preservation Constraints

Key points

• We consider improving a design while holding selected state observables near trusted values, which we call state preservation constraints.1
• Constrained Bayesian optimisation handles these with a learnt feasibility model, but struggles with this problem's highly anisotropic feasible set.1
• 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.1
• We demonstrate the method end-to-end on our key application - Tokamak divertor optimisation under plasma-boundary preservation.1

Sources (1)

• [1]Bayesian Optimisation under State-Preservation Constraints
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 03:34 PM

  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.

Extractive summary: sentences quoted from the sources.