Bayesian optimization with Gaussian-process-based active machine learning for projection multi-photon 3D printing

The rapidly developing frontiers of additive manufacturing, especially multi-photon lithography, create a constant need for optimization of new process parameters. The recently developed projection multi-photon lithography process used in this work is one such example. This work presents an active machine learning framework which can serve as a guide for exploration of these uncharted parameter spaces. The framework uses Bayesian optimization to guide experimentation to dynamically collect the most optimal data for training of a Gaussian process regression machine learning model. This model then serves as a surrogate for the manufacturing process by predicting optimal process parameters for printing of a target geometry. The results of the framework for several 2D shapes are shown and the extension of this framework to 3D structures is discussed.