Physics‐informed Transfer Learning for Out‐of‐sample Vapor Pressure Predictions
Authors:
- 27 February 2023
- journal article
- research article
- Published by Wiley in Propellants, Explosives, Pyrotechnics
- Vol. 48 (4)
Abstract
Recent advances have enabled machine learning methodologies developed for large datasets to be applied to the small experimental datasets typically available for chemical systems. Such advances typically involve a data‐based approach to transfer learning, where a portion of the experimental data for the property of interest is used to fine‐tune a model that is pre‐trained on computationally generated data. This transfer learning approach does not work for very small experimental datasets, where there are only enough data for model validation. Here, we develop a physics‐informed transfer learning strategy to train a directed‐message passing neural network (D‐MPNN) model, enabling extrapolation outside of the training domain. We demonstrate this approach by training a D‐MPNN model on interpolated vapor pressures and validate the model on an out‐of‐sample test set of energetic molecule vapor pressures, achieving accuracy comparable to those of experiments.Keywords
Funding Information
- Defense Advanced Research Projects Agency
This publication has 81 references indexed in Scilit:
- Kinetics and enthalpy of nitroglycerin evaporation from double base propellants by isothermal thermogravimetryThermochimica Acta, 2010
- Gas Saturation Vapor Pressure Measurements of Mononitrotoluene Isomers from (283.15 to 313.15) KJournal of Chemical & Engineering Data, 2009
- Vapor Pressures of 1-Methyl-2-pyrrolidone, 1-Methyl-azepan-2-one, and 1,2-Epoxy-3-chloropropaneJournal of Chemical & Engineering Data, 2007
- An Improved Force Field for the Prediction of the Vapor−Liquid Equilibria for Carboxylic AcidsThe Journal of Physical Chemistry B, 2004
- Prediction of Vapor Pressures of Hydrocarbons and Halohydrocarbons from Molecular Structure with a Computational Neural Network ModelJournal of Chemical Information and Computer Sciences, 1999
- Saturated Vapor Pressure Measurements on Isomeric Mononitrotoluenes at Temperatures between 380 and 460 KJournal of Chemical & Engineering Data, 1994
- A neural network model extracting features from speech signalsSystems and Computers in Japan, 1988
- Do we understand the physics in the constitutive equation?Journal of Non-Newtonian Fluid Mechanics, 1988
- Measurement of .gamma..infin. using GLC. 2. Results for the stationary phases N-formylmorpholine and N-methylpyrrolidoneJournal of Chemical & Engineering Data, 1987
- Vapor-liquid equilibrium data for binary systems of aniline with acetone, acetonitrile, chlorobenzene, methanol, and 1-penteneJournal of Chemical & Engineering Data, 1980