Machine learning-predicted quantum control landscape maps of laser-induced three-dimensional alignment of asymmetric-top molecules

Kawaoto, Ryoma*; Namba, Tomotaro  ; Kumagai, Yuta   ; Otsuki, Yukiyoshi*

Machine learning (ML)-based approaches are adopted to predict quantum control landscape maps of laser-induced three-dimensional alignment of asymmetric-top molecules. The landscape map of each molecule is composed of thousands of pixels, each of which shows the maximum degree of alignment under specified set of control parameters. By overcoming the difficulties originating from the considerably different molecular parameters, we develop the ML models to successfully predict the maps, whereby a wide view of alignment control is provided.