Efficient Graph Neural Networks for Predicting Chemistry

Graph Neural Networks (GNNs) are powerful deep learning models for predicting molecular properties but they require large datasets and significant computational resources. My research focuses on enhancing their efficiency and accuracy by improving how these models learn and process information internally. By inventing new model architectures and training strategies, I demonstrate that GNNs can achieve superior prediction accuracy while using either the same or even fewer computational resources. My work advances the development of efficient GNNs for chemistry prediction.