DeepEOSNet: Capturing the dependency on thermodynamic state in property prediction tasks

We propose a machine learning (ML) architecture to better capture the dependency of thermodynamic properties on independent states. When predicting state-dependent thermodynamic properties, ML models need to account for both molecular structure and the thermodynamic state, described by independent variables, typically temperature, pressure, and composition. Modern molecular ML models include state information by adding it to molecular fingerprint vectors or by embedding explicit (semi-empirical) thermodynamic relations. We propose splitting the information processing on the molecular structure and the dependency on states into two separate network channels: a graph neural network and a multilayer perceptron, whose output is combined by a dot product. We refer to our approach as DeepEOSNet, as it is based on the DeepONet architecture (Lu et al., 2021): instead of operators, we learn state dependencies, with the possibility to predict equation of states (EOS). We investigate the predictive performance of DeepEOSNet in three case studies, including the prediction of vapor pressure as a function of temperature, and mixture molar volume as a function of composition, temperature, and pressure. Our results show superior performance of DeepEOSNet for predicting vapor pressure and comparable performance for predicting mixture molar volume compared to state-of-research graph-based thermodynamic prediction models from our earlier works. In fact, we see large potential of DeepEOSNet when data is sparse in the state domain and the output function is structurally similar across different molecules. The concept of DeepEOSNet can easily be transferred to other deep learning architectures, providing a viable option for molecular property prediction.