Predicting adsorption energies and the physical properties of H, N, and O adsorbed on transition metal surfaces: A machine learning study

•Adsorption energy, adsorption height, and buckling of the surface was predicted utilizing machine learning, specifically the hierarchically interacting particle neural network (HIP-NN), for H, N, and O on bimetallic transition metal surfaces.•These adsorption properties were predicted employing clean surface geometries.•Including periodic boundary conditions into the machine learning algorithm improved results.•Buckling of surface, as predicted by a machine learning mode, was successfully employed to prune the training dataset.