Numerical and experimental generated data during project https://doi.org/10.1038/s41598-024-82701-3

The dataset was generated using a procedure for expedited globalized parameter adjustment of microwave passives. The search process was embedded in a surrogate-assisted machine learning framework operating in a dimensionality-restricted domain, spanned by the parameter space directions being of importance in terms of their effects on the circuit characteristic variability. Extensive comparisons with several state-of-the-art routines, including a bio-inspired algorithm and an ML scheme operating within the original parameter space, indicated competitive efficacy regarding the quality of the rendered designs and CPU efficiency. The CPU savings achieved due to dimensionality reduction were as high as 50%.