Phys.org  October 25, 2024 Fast and accurate prediction of dielectric function facilitates the development of novel dielectric materials, an ingredient of many cutting-edge technologies such as 6G networks. Researchers in Japan introduced a versatile machine-learning scheme implemented in Git hub for predicting dipole moments of molecular liquids to study dielectric properties. They attributed the center of mass of Wannier functions (called Wannier centers), to each chemical bond and created neural network models. They applied liquid methanol and ethanol to demonstrate that their neural network models successfully predicted the dipole moment of various liquid configurations in close agreement with DFT calculations. […]