Computer Science and Engineering assistant professor Dr. Mali has published a paper in Nature Communications titled “Using a physics-informed neural network and fault zone acoustic monitoring to predict lab earthquakes” (reference: Nat Commun 14, 3693 (2023).) The paper describes a machine learning based method for predicting laboratory earthquakes using micro-failure events and temporal evolution of fault zone elastic properties. The results come from purely data-driven models trained with large datasets. The work uses a novel Physics-Informed Neural Network (PINN) model.
This ground breaking work offers a promising path for predicting failures using machine learning techniques and thus improving our understanding of earthquakes. The ability to predict earthquakes will have major societal impact. The work was completed and published with five colleagues from Pennsylvania State University.