Hybridizing Physical and Data-driven Prediction Methods for Physicochemical Properties
This work addresses the challenge of accurate property prediction in chemistry, though it appears incremental as it builds on existing hybrid approaches.
The paper tackles the problem of predicting physicochemical properties by hybridizing physical and data-driven methods, achieving significant improvements over baselines and established ensemble methods.
We present a generic way to hybridize physical and data-driven methods for predicting physicochemical properties. The approach `distills' the physical method's predictions into a prior model and combines it with sparse experimental data using Bayesian inference. We apply the new approach to predict activity coefficients at infinite dilution and obtain significant improvements compared to the data-driven and physical baselines and established ensemble methods from the machine learning literature.