Abstract:Accurately predicting activation energies is crucial for understanding chemical reactions and modeling complex reaction systems. However, the high computational cost of quantum chemistry methods often limits the feasibility of large-scale studies, leading to a scarcity of high-quality activation energy data. In this work, we explore and compare three innovative approaches—transfer learning, delta learning, and feature engineering—to enhance the accuracy of activation energy predictions using graph neural networks, specifically focusing on methods that incorporate low-cost, low-level computational data. Using the Chemprop model, we systematically evaluated how these methods leverage data from semiempirical quantum mechanical (SQM) calculations to improve predictions. Our findings reveal that delta learning, which adjusts low-level SQM activation energies to match high-level CCSD(T)-F12a targets, offers the greatest accuracy improvement. However, this approach also requires more computational resources due to the complexity of transition state searches for each query of activation energy during model application. Transfer learning, which pretrains models on large datasets of low-level data, provided mixed results, particularly when there was a mismatch in the reaction distributions between the training and target datasets. Feature engineering, which involves adding computed molecular properties as input features, showed modest gains, particularly when incorporating thermodynamic properties. Our study highlights the trade-offs between accuracy and computational demand in selecting the best approach for enhancing activation energy predictions. These insights provide valuable guidelines for researchers aiming to apply machine learning in chemical reaction engineering, helping to balance accuracy with resource constraints.

When Do Quantum Mechanical Descriptors Help Graph Neural Networks to Predict Chemical Properties?

When Do Quantum Mechanical Descriptors Help Graph Neural Networks Predict Chemical Properties?

Quantum chemistry-augmented neural networks for reactivity prediction: Performance, generalizability, and explainability

Toward Accurate Predictions of Atomic Properties via Quantum Mechanics Descriptors Augmented Graph Convolutional Neural Network: Application of This Novel Approach in NMR Chemical Shifts Predictions

Extreme Acceleration of Graph Neural Network-based Prediction Models for Quantum Chemistry

Universal neural network potentials as descriptors: Towards scalable chemical property prediction using quantum and classical computers

Introducing Block Design in Graph Neural Networks for Molecular Properties Prediction

Molecular Property Prediction: A Multilevel Quantum Interactions Modeling Perspective

Hybrid Quantum Graph Neural Network for Molecular Property Prediction

Quantum-chemical insights from deep tensor neural networks

Quantum Graph Neural Network Models for Materials Search

Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models

Fast Quantum Property Prediction via Deeper 2D and 3D Graph Networks

Enhancing Activation Energy Predictions under Data Constraints Using Graph Neural Networks

Cross-Modal Prediction of Spectral and Structural Descriptors via a Pretrained Model Enhanced with Chemical Insights

Chemical Properties from Graph Neural Network-Predicted Electron Densities

Quantum Mechanics and Machine Learning Synergies: Graph Attention Neural Networks to Predict Chemical Reactivity

AI-enhanced chemical paradigm: From molecular graphs to accurate prediction and mechanism

Directed graph attention neural network utilizing 3D coordinates for molecular property prediction

Quantum-Chemical Insights from Interpretable Atomistic Neural Networks

Heterogeneous Molecular Graph Neural Networks for Predicting Molecule Properties