The high‐entropy strategy has gained prominence for enhancing the energy storage density and efficiency of lead‐free dielectric capacitors. However, exploring high‐performance ceramics within the vast compositional space of high‐entropy systems remains challenging using traditional trial‐and‐error approaches. In this study, a random forest regression model was developed using a dataset of BaTiO 3 (BT)‐based ceramics, incorporating four key descriptors: configurational entropy (Δ S ), tolerance factor ( t ), Pauling electronegativity (PEG), and valence electron concentration (VEC/Z). An expected improvement acquisition function guided iterative experimentation across a compositional space of 110 000 candidates. The synergistic effects of multiphase coexistence and nanoscale polar clusters enable the maintenance of high polarization and minimal hysteresis under applied electric fields. As a result, a superparaelectric relaxor ferroelectric (SPE‐RFE) ceramic was achieved, exhibiting a high energy storage density of 8.91 J cm −3 and an efficiency of 92.98%. Furthermore, excellent charge–discharge performance, along with temperature and frequency stability, underscores its potential for practical applications, demonstrating the efficacy of machine learning in advancing energy storage ceramics.
Liu et al. (Tue,) studied this question.