Combined first--principles calculation and neural--network correction approach as a powerful tool in computational physics and chemistry

摘要

Despite of their success, the results of first-principles quantum mechanicalcalculations contain inherent numerical errors caused by variousapproximations. We propose here a neural-network algorithm to greatly reducethese inherent errors. As a demonstration, this combined quantum mechanicalcalculation and neural-network correction approach is applied to the evaluationof standard heat of formation $\DelH$ and standard Gibbs energy of formation$\DelG$ for 180 organic molecules at 298 K. A dramatic reduction of numericalerrors is clearly shown with systematic deviations being eliminated. Forexamples, the root--mean--square deviation of the calculated $\DelH$ ($\DelG$)for the 180 molecules is reduced from 21.4 (22.3) kcal$\cdotp$mol$^{-1}$ to 3.1(3.3) kcal$\cdotp$mol$^{-1}$ for B3LYP/6-311+G({\it d,p}) and from 12.0 (12.9)kcal$\cdotp$mol$^{-1}$ to 3.3 (3.4) kcal$\cdotp$mol$^{-1}$ forB3LYP/6-311+G(3{\it df},2{\it p}) before and after the neural-networkcorrection.