Transfer Learning in Bayesian Optimization for Aircraft Design
This work addresses optimization challenges in aircraft design, but it is incremental as it builds on existing transfer learning and Bayesian optimization methods.
The paper tackled the cold start problem in Bayesian optimization for aircraft design by using transfer learning with an ensemble of surrogate models, showing significant improvement in early convergence and prediction accuracy for objectives and constraints.
The use of transfer learning within Bayesian optimization addresses the disadvantages of the so-called \textit{cold start} problem by using source data to aid in the optimization of a target problem. We present a method that leverages an ensemble of surrogate models using transfer learning and integrates it in a constrained Bayesian optimization framework. We identify challenges particular to aircraft design optimization related to heterogeneous design variables and constraints. We propose the use of a partial-least-squares dimension reduction algorithm to address design space heterogeneity, and a \textit{meta} data surrogate selection method to address constraint heterogeneity. Numerical benchmark problems and an aircraft conceptual design optimization problem are used to demonstrate the proposed methods. Results show significant improvement in convergence in early optimization iterations compared to standard Bayesian optimization, with improved prediction accuracy for both objective and constraint surrogate models.